Coal mine support safety risk prediction and early warning system based on big data analysis

Through big data analysis and machine learning technology, a coal mine support safety risk prediction and early warning system has been built, which has solved the problems of low monitoring efficiency and lack of risk prediction in the existing technology, realized automated and intelligent security risk monitoring, and reduced missed reports and risk of false alarms.

CN118917489APending Publication Date: 2024-11-08SHANXI JINGUANTONG METAL TECHNOLOGY CO LTD

Patent Information

Application Number
CN202411123015.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The existing coal mine support monitoring methods are inefficient and cannot achieve full coverage. They rely on empiricism and qualitative analysis, lack scientific quantitative models and risk prediction capabilities, and are prone to missed or false positives, and fail to achieve risk prediction results.

Method used

Adopt a coal mine support safety risk prediction and early warning system based on big data analysis, including historical data collection module, structural model training module, risk model training module and risk monitoring module, and train time series models and risk states through historical data of environmental and structural parameters. Model to realize future trend forecasting and security risk warning of structural parameters.

Benefits of technology

The automation and intelligence level of coal mine support safety monitoring has been improved, manual participation has been reduced, the risks of missed and false alarms have been reduced, and real-time security risk warning and prediction have been achieved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118917489A_ABST
    Figure CN118917489A_ABST
Patent Text Reader

Abstract

The invention discloses a coal mine support safety risk prediction and early warning system based on big data analysis, and relates to the technical field of safety early warning. A historical environment sequence, a historical structure sequence and historical risk time are collected by setting a historical data collection module, and a future value prediction model is trained by setting a structure model training module; setting a risk model training module to train a risk state prediction model, setting a risk monitoring module to collect an initial environment sequence and an initial structure sequence of a to-be-monitored coal mine support, constructing a structure prediction instance, obtaining a future prediction value of each structure parameter, and constructing a time prediction instance based on the future prediction value of each structure parameter. Based on the time prediction instance and the risk state prediction model, initiating security risk early warning; the automation level and the intelligence level of coal mine support safety monitoring are effectively improved, manual participation is reduced, and the risks of missing report and false report are reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of safety early warning technology, and in particular to a coal mine support safety risk prediction and early warning system based on big data analysis. Background Art

[0002] Coal mining is a high-risk operation. The underground environment is extremely harsh and changeable. The stability of the support structure is crucial to ensure the safe operation of the entire mine. However, during long-term use, the support structure faces many severe challenges and damage risks:

[0003] As the mining face advances, the stress state of the surrounding rock is constantly redistributed, and the original support structure may exceed the designed bearing capacity and become unstable and damaged. Rock strata of different properties will also produce large deformations and displacements under the action of mining unloading, causing problems such as deformation and cracking of the support structure and rock slag peeling. In addition, if the mining area encounters complex geological structures such as faults and fracture zones, it will exert greater squeezing or shearing forces on the support, increasing its risk of instability.

[0004] Therefore, how to evaluate the health status of the support structure in real time, accurately and comprehensively is crucial to avoid major safety accidents and protect the lives and property of mine personnel. However, there are still many shortcomings in the existing support monitoring methods: the traditional method mainly relies on regular on-site manual inspections, which is inefficient and cannot achieve full coverage; although local measuring points such as strain gauges and displacement meters are also deployed, they can only reflect point information and it is difficult to establish a dynamic assessment model for the overall health of the support; in addition, the existing methods rely too much on empiricism and qualitative analysis, lack scientific quantitative models and risk prediction capabilities, and are prone to omissions or false alarms.

[0005] The Chinese patent with the authorization announcement number CN114215564B discloses an intelligent early warning method for coal mine support equipment. The early warning is carried out by using a monitoring module, a storage module and an early warning system. The monitoring module includes a pressure monitoring module and a positioning module. The pressure monitoring module monitors the working resistance of the hydraulic support of the working face in real time. The positioning module monitors the position of the hydraulic support of the working face. The storage module receives the information sent by the monitoring module and sends the information to the early warning system and the remote server. The early warning system receives the information sent by the storage module and processes and analyzes it. According to the statistical data of mine mining practice, the pressure early warning threshold of a single working face hydraulic support and the total pressure early warning threshold of a group of adjacent working face hydraulic supports are determined. The early warning reminder is issued by judging whether the actual pressure value is greater than the early warning threshold. However, this method is based on early warning based on real-time data and fails to achieve the effect of risk prediction.

[0006] To this end, the present invention proposes a coal mine support safety risk prediction and early warning system based on big data analysis. Summary of the invention

[0007] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a coal mine support safety risk prediction and early warning system based on big data analysis, which effectively improves the automation and intelligence level of coal mine support safety monitoring, reduces manual participation, and reduces the risk of missed reports and false reports.

[0008] In order to achieve the above purpose, a coal mine support safety risk prediction and early warning system based on big data analysis is proposed, which includes a historical data collection module, a structural model training module, a risk model training module and a risk monitoring module; wherein each module is electrically connected;

[0009] A historical data collection module collects in advance the historical environmental sequences of various environmental parameters of each tested coal mine support during the test process, the historical structural sequences of various structural parameters changing over time, and the historical risk time of safety risks; the historical data collection module sends the historical environmental sequence and historical structural sequence to the structural model training module; and sends the historical environmental sequence, historical structural sequence and historical risk time to the risk model training module;

[0010] The structural model training module constructs structural prediction samples for various structural parameters based on the historical environment sequence and historical structure sequence of each test coal mine support, and trains future value prediction models for various structural parameters based on the structural prediction samples; the structural model training module sends the trained future value prediction model to the risk monitoring module;

[0011] The risk model training module constructs a time prediction sample based on the historical structure sequence and historical risk time of each test coal mine support, and trains a risk state prediction model based on the time prediction sample; the risk model training module sends the trained risk state prediction model to the risk monitoring module;

[0012] The risk monitoring module collects the initial environment sequence of the coal mine support to be monitored and the initial structure sequence of each structural parameter, constructs a structure prediction instance based on the initial environment sequence and the initial structure sequence, obtains the future prediction value of each structural parameter based on the structure prediction instance and the future value prediction model, constructs a time prediction instance based on the future prediction value of each structural parameter, and initiates a safety risk warning based on the time prediction instance and the risk status prediction model; and updates the initial environment sequence and the initial structure sequence over time, cyclically constructs the structure prediction instance, the time prediction instance and the safety risk warning; and updates the initial environment sequence and the initial structure sequence over time, cyclically constructs the structure prediction instance, the time prediction instance and the safety risk warning.

[0013] Specifically, the collection of the historical environment sequence, the historical structure sequence of changes of various structural parameters over time, and the historical risk time includes the following steps:

[0014] Step 11: Arrange environmental parameter sensors for various environmental parameters in the area where the test coal mine support installation position is located, and collect parameter values ​​of various environmental parameters in real time through the environmental parameter sensors; the parameter sequence of various environmental parameters arranged in chronological order is regarded as a set of historical environmental sequences;

[0015] Step 12: Install structural parameter sensors corresponding to various structural parameters at key support positions of the test coal mine support, and collect parameter values ​​of various structural parameters in real time through various structural parameter sensors; the key support positions are manually selected in advance according to the physical structural characteristics of the coal mine support structure; the structural sequence of various structural parameters arranged in chronological order is taken as a set of historical structural sequences;

[0016] Step 13: Coal mine site personnel record the time when safety problems occur in each coal mine support structure as the historical risk time.

[0017] Specifically, the method of constructing a structural prediction sample for each structural parameter based on the historical environment sequence and historical structure sequence of each test coal mine support is as follows:

[0018] Preset prediction time length t;

[0019] Preset the time window length w and sliding step size s;

[0020] For each structural parameter of each tested coal mine support:

[0021] With a time window length w and a sliding step length s, a sliding window method is used to obtain environmental subsequences and structural subsequences with a length of the time window length w from each historical environmental sequence and the historical structural sequence of the corresponding structural parameters. The environmental subsequences and structural subsequences obtained at each sliding constitute the structural feature samples.

[0022] The parameter value of the structural parameter corresponding to the moment of prediction time length t after each sliding moment in the historical structural sequence is used as the structural label sample;

[0023] Each group of structural feature samples and structural label samples constitutes a structural prediction sample.

[0024] Specifically, the method of training the future value prediction model for each structural parameter based on the structural prediction sample is as follows:

[0025] For each structural parameter:

[0026] The structural feature sample of each structural prediction sample is used as the input of the future value prediction model, and the future value prediction model takes the predicted value of the structural parameter after the prediction time length t as the output; the future value prediction model takes the structural label sample corresponding to the structural prediction sample as the prediction target, and takes the difference between the predicted value of the structural parameter and the structural label sample as the first prediction error, and takes minimizing the sum of squares of the first prediction errors of all structural prediction samples as the training target; the future value prediction model is trained until the sum of squares of the first prediction error reaches convergence and the training is stopped; the future value prediction model is any one of the time series prediction models.

[0027] Specifically, the method of constructing the time prediction sample based on the historical structure sequence and historical risk time of each test coal mine support is:

[0028] For each tested coal mine support:

[0029] The parameter values ​​of each structural parameter at each moment are combined into a set of time prediction vectors;

[0030] A risk status label is set for each moment; the risk status label is set in such a way that if the moment is a historical risk time when a security risk occurs, the risk status label is set to 1; if the moment is not a historical risk time when a security risk occurs, the risk status label is set to 0;

[0031] The time prediction vector and risk status label at each moment constitute the time prediction sample.

[0032] Specifically, the method of training the risk status prediction model based on the time prediction sample is as follows:

[0033] The time prediction vector of each test coal mine support at each moment is used as the input of the risk state prediction model, and the risk state prediction model takes the predicted value of the risk state at each moment as the output, and the predicted value of the risk state is 0 or 1; the risk state prediction model takes the risk state label corresponding to each moment as the prediction target, and the difference between the predicted value of the risk state and the risk state label as the second prediction error, and takes minimizing the sum of squares of the second prediction errors at all moments as the training target; the risk state prediction model is trained until the sum of squares of the second prediction errors converges; the risk state prediction model is stopped; the risk state prediction model is any one of the classification models.

[0034] Specifically, the method of obtaining the future prediction value of each structural parameter based on the structural prediction instance and the future value prediction model is as follows:

[0035] For each structural parameter, the structural prediction instance is input into the future value prediction model, and the predicted value of the structural parameter after the prediction time length t output by the future value prediction model is obtained as the future prediction value;

[0036] The method of constructing a time prediction instance based on the future prediction values ​​of each structural parameter and initiating a safety risk warning based on the time prediction instance and the risk status prediction model is as follows:

[0037] The future prediction values ​​of each structural parameter are combined into a set of structural prediction value vectors as a set of time prediction instances;

[0038] The structural prediction value vector is input into the risk state prediction model to obtain the predicted value of the risk state after the prediction time length t output by the risk state prediction model. If the predicted value of the risk state is 1, a safety risk alarm is initiated; otherwise, no safety risk alarm is initiated.

[0039] A coal mine support safety risk prediction and early warning method based on big data analysis is proposed, which includes the following steps:

[0040] Step 1: Collect in advance the historical environmental sequences of various environmental parameters of each test coal mine support during the test process, the historical structural sequences of various structural parameters changing over time, and the historical risk time of safety risks;

[0041] Step 2: Based on the historical environment sequence and historical structure sequence of each test coal mine support, construct a structural prediction sample for each structural parameter, and based on the structural prediction sample, train a future value prediction model for each structural parameter;

[0042] Step 3: Based on the historical structure sequence and historical risk time of each test coal mine support, construct a time prediction sample, and train a risk state prediction model based on the time prediction sample;

[0043] Step 4: Collect the initial environment sequence of the coal mine support to be monitored and the initial structure sequence of each structure parameter;

[0044] Step 5: Based on the initial environment sequence and the initial structure sequence, a structure prediction instance is constructed, and based on the structure prediction instance and the future value prediction model, the future prediction values ​​of each structure parameter are obtained;

[0045] Step 6: Based on the future predicted values ​​of each structural parameter, a time prediction instance is constructed, and based on the time prediction instance and the risk status prediction model, a safety risk warning is initiated;

[0046] Step 7: Update the initial environment sequence and the initial structure sequence over time, and execute steps 5 to 6 repeatedly.

[0047] An electronic device is proposed, comprising: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0048] The processor executes the above-mentioned coal mine support safety risk prediction and early warning method based on big data analysis by calling the computer program stored in the memory.

[0049] A computer-readable storage medium is provided, on which a rewritable computer program is stored;

[0050] When the computer program runs on a computer device, the computer device executes the above-mentioned coal mine support safety risk prediction and early warning method based on big data analysis.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] The present invention sets a historical data collection module to pre-collect the historical environment sequences of various environmental parameters of each test coal mine support during the test process, the historical structure sequences of various structural parameters changing with time, and the historical risk time of safety risks; sets a structure model training module to construct structure prediction samples for various structural parameters based on the historical environment sequences and historical structure sequences of each test coal mine support, and trains future value prediction models for various structural parameters based on the structure prediction samples; sets a risk model training module to construct time prediction samples based on the historical structure sequences and historical risk times of each test coal mine support, and trains risk state prediction models based on the time prediction samples; sets a risk monitoring module to collect the initial environment sequence of the coal mine support to be monitored and the initial structure sequence of various structural parameters, constructs a structure prediction instance based on the initial environment sequence and the initial structure sequence, obtains the future prediction value of each structural parameter based on the structure prediction instance and the future value prediction model, constructs a time prediction instance based on the future prediction value of each structural parameter, and initiates a safety risk warning based on the time prediction instance and the risk state prediction model; and updates the initial environment sequence and the initial structure sequence over time, and cyclically constructs the structure prediction instance, the time prediction instance, and the safety risk warning; and fully utilizes the advantages of big data and machine learning technology. First, by deploying a fully covered online monitoring network, high-quality time series data of environmental and structural parameters can be collected; second, by training time series models based on these historical data, the future parameter change trends of the support can be predicted, and the risk status model can be trained to predict the time of future safety accidents; third, the data of the support to be tested can be received in real time and future situations can be cyclically predicted, so as to issue early warnings; this effectively improves the automation and intelligence level of monitoring, reduces manual participation, and reduces the risk of missed reports and false alarms. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1This is a module connection relationship diagram of the coal mine support safety risk prediction and early warning system based on big data analysis in Example 1 of the present invention;

[0054] Figure 2 This is a flow chart of a method for predicting and warning coal mine support safety risks based on big data analysis in Example 2 of the present invention;

[0055] Figure 3 This is a schematic diagram of the structure of an electronic device in Embodiment 3 of the present invention;

[0056] Figure 4 This is a schematic diagram of the computer-readable storage medium structure in Example 4 of the present invention. DETAILED DESCRIPTION

[0057] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] Example 1

[0059] like Figure 1 As shown, the coal mine support safety risk prediction and early warning system based on big data analysis includes a historical data collection module, a structural model training module, a risk model training module and a risk monitoring module; wherein each module is electrically connected;

[0060] The historical data collection module collects in advance the historical environmental sequence of various environmental parameters of each tested coal mine support during the test process, the historical structural sequence of various structural parameters changing over time, and the historical risk time when safety risks occurred;

[0061] The test coal mine supports are several coal mine supports pre-selected as test coal mine supports, so as to conduct experiments through the test coal mine supports to obtain sample data for model training;

[0062] The structural parameters of the tested coal mine support are key parameters that can reflect and affect the safety risk of the coal mine support, including but not limited to strain, displacement and cracks;

[0063] The environmental parameters include but are not limited to temperature, humidity, environmental stress, etc.

[0064] The collection of the historical environment sequence, the historical structure sequence of changes of various structural parameters over time, and the historical risk time includes the following steps:

[0065] Step 11: Arrange environmental parameter sensors for various environmental parameters in the area where the test coal mine support installation position is located, and collect parameter values ​​of various environmental parameters in real time through the environmental parameter sensors; the parameter sequence of various environmental parameters arranged in chronological order is regarded as a set of historical environmental sequences;

[0066] Step 12: Install structural parameter sensors corresponding to various structural parameters at key support positions of the test coal mine support, and collect parameter values ​​of various structural parameters in real time through various structural parameter sensors; for example, the structural parameter sensors include strain sensors, displacement sensors and image capture devices (identifying the number and length of cracks on the coal mine support structure through image processing technology); the key support positions are manually selected in advance according to the physical structural characteristics of the coal mine support structure; the structural sequence of various structural parameters arranged in chronological order is taken as a group of historical structural sequences;

[0067] Step 13: Coal mine site personnel record the time when safety problems occur in each coal mine support structure as the historical risk time.

[0068] The historical data collection module sends the historical environment sequence and the historical structure sequence to the structure model training module; and sends the historical environment sequence, the historical structure sequence and the historical risk time to the risk model training module.

[0069] The structural model training module constructs structural prediction samples for various structural parameters based on the historical environmental sequences and historical structural sequences of each test coal mine support, and trains future value prediction models for various structural parameters based on the structural prediction samples;

[0070] Furthermore, the method of constructing a structural prediction sample for each structural parameter based on the historical environment sequence and historical structure sequence of each test coal mine support is as follows:

[0071] Preset prediction time length t;

[0072] Preset the time window length w and sliding step size s;

[0073] For each structural parameter of each tested coal mine support:

[0074] With a time window length w and a sliding step length s, a sliding window method is used to obtain environmental subsequences and structural subsequences with a length of the time window length w from each historical environmental sequence and the historical structural sequence of the corresponding structural parameters. The environmental subsequences and structural subsequences obtained at each sliding constitute the structural feature samples.

[0075] The parameter value of the structural parameter corresponding to the moment of prediction time length t after each sliding moment in the historical structural sequence is used as the structural label sample;

[0076] Each group of structural feature samples and structural label samples constitutes a structural prediction sample;

[0077] Specifically, the sliding window method is a conventional technical means in the art, and a simple example is used here to illustrate:

[0078] The preset prediction time length t is 1 hour, the preset time window length w is 4, and the preset sliding step length s is 1;

[0079] The historical environment sequence and historical structure sequence collected in chronological order are:

[0080] Temperature: [25.2, 25.6, 26.1, 26.7, 27.2, 29.1];

[0081] Humidity: [62%, 60%, 58%, 56%, 55%, 51%];

[0082] strain: [0.116, 0.122, 0.129, 0.135, 0.142, 0.173];

[0083] Then two structure prediction samples can be constructed by the sliding window method:

[0084] Structural feature sample 1:

[0085] Temperature: [25.2, 25.6, 26.1, 26.7], Humidity: [62%, 60%, 58%, 56%], Strain: [0.116, 0.122, 0.129, 0.135];

[0086] Structure label sample 1: strain label: [0.142];

[0087] Structural feature sample 2: temperature: [25.6, 26.1, 26.7, 27.2], humidity: [60%, 58%, 56%, 55%], strain: [0.122, 0.129, 0.135, 0.142];

[0088] Structure label sample 2: Strain label: [0.173].

[0089] Furthermore, the method of training the future value prediction model for each structural parameter based on the structural prediction sample is as follows:

[0090] For each structural parameter:

[0091] The structural feature sample of each structural prediction sample is used as the input of the future value prediction model, and the future value prediction model takes the predicted value of the structural parameter after the prediction time length t as the output; the future value prediction model takes the structural label sample corresponding to the structural prediction sample as the prediction target, takes the difference between the predicted value of the structural parameter and the structural label sample as the first prediction error, and takes minimizing the sum of squares of the first prediction errors of all structural prediction samples as the training target; the future value prediction model is trained until the sum of squares of the first prediction error reaches convergence and the training is stopped; the future value prediction model is any one of the time series prediction models, and the time series prediction model includes but is not limited to the RNN model or the LSTM model, etc.;

[0092] The structural model training module sends the trained future value prediction model to the risk monitoring module.

[0093] The risk model training module builds time prediction samples based on the historical structure sequence and historical risk time of each test coal mine support, and trains the risk status prediction model based on the time prediction samples;

[0094] The method of constructing the time prediction sample based on the historical structure sequence and historical risk time of each test coal mine support is as follows:

[0095] For each tested coal mine support:

[0096] The parameter values ​​of each structural parameter at each moment are combined into a set of time prediction vectors;

[0097] A risk status label is set for each moment; the risk status label is set in such a way that if the moment is a historical risk time when a security risk occurs, the risk status label is set to 1; if the moment is not a historical risk time when a security risk occurs, the risk status label is set to 0;

[0098] The time prediction vector and risk status label at each moment constitute the time prediction sample.

[0099] The method of training the risk status prediction model based on the time prediction sample is as follows:

[0100] The time prediction vector of each test coal mine support at each moment is used as the input of the risk state prediction model, and the risk state prediction model takes the predicted value of the risk state at each moment as the output, and the predicted value of the risk state is 0 or 1; the risk state prediction model takes the risk state label corresponding to each moment as the prediction target, takes the difference between the predicted value of the risk state and the risk state label as the second prediction error, and takes minimizing the sum of squares of the second prediction errors at all moments as the training target; the risk state prediction model is trained until the sum of squares of the second prediction errors reaches convergence, and the training is stopped; the risk state prediction model is any one of the classification models, and the classification model includes but is not limited to a DNN model or an SVM model, etc.;

[0101] The risk model training module sends the trained risk status prediction model to the risk monitoring module.

[0102] The risk monitoring module collects the initial environment sequence of the coal mine support to be monitored and the initial structure sequence of each structural parameter, constructs a structure prediction instance based on the initial environment sequence and the initial structure sequence, obtains the future prediction value of each structural parameter based on the structure prediction instance and the future value prediction model, constructs a time prediction instance based on the future prediction value of each structural parameter, and initiates a safety risk warning based on the time prediction instance and the risk state prediction model; and updates the initial environment sequence and the initial structure sequence over time, cyclically constructing the structure prediction instance, the time prediction instance and the safety risk warning;

[0103] The method of collecting the initial environment sequence of the coal mine support to be monitored and the initial structure sequence of each structure parameter is as follows:

[0104] For each environmental parameter of the coal mine support to be monitored, the parameter values ​​of the time window length w are continuously collected to form an initial environmental sequence;

[0105] For each structural parameter of the coal mine support to be monitored, the parameter values ​​of the time window length w are continuously collected to form an initial structural sequence;

[0106] Furthermore, the method of constructing a structure prediction instance based on the initial environment sequence and the initial structure sequence is:

[0107] For each structural parameter, the initial environment sequence and the initial structure sequence corresponding to the structural parameter are combined into a group of structure prediction instances.

[0108] The method of obtaining the future prediction value of each structural parameter based on the structural prediction instance and the future value prediction model is as follows:

[0109] For each structural parameter, the structural prediction instance is input into the future value prediction model, and the predicted value of the structural parameter after the prediction time length t output by the future value prediction model is obtained as the future prediction value;

[0110] The method of constructing a time prediction instance based on the future prediction values ​​of each structural parameter and initiating a safety risk warning based on the time prediction instance and the risk status prediction model is as follows:

[0111] The future prediction values ​​of each structural parameter are combined into a set of structural prediction value vectors as a set of time prediction instances;

[0112] The structural prediction value vector is input into the risk state prediction model to obtain the predicted value of the risk state after the prediction time length t output by the risk state prediction model. If the predicted value of the risk state is 1, a safety risk alarm is initiated; otherwise, no safety risk alarm is initiated.

[0113] The method of updating the initial environment sequence and the initial structure sequence over time is:

[0114] Using the sliding window method, with the sliding step length s as the sliding compensation interval, the parameter values ​​of various environmental parameters and structural parameters of the coal mine support to be monitored are continuously collected to form the initial environmental sequence and initial structural sequence of the time window length w respectively.

[0115] Example 2

[0116] like Figure 2 As shown in the figure, the coal mine support safety risk prediction and early warning method based on big data analysis includes the following steps:

[0117] Step 1: Collect in advance the historical environmental sequences of various environmental parameters of each test coal mine support during the test process, the historical structural sequences of various structural parameters changing over time, and the historical risk time of safety risks;

[0118] Step 2: Based on the historical environment sequence and historical structure sequence of each test coal mine support, construct a structural prediction sample for each structural parameter, and based on the structural prediction sample, train a future value prediction model for each structural parameter;

[0119] Step 3: Based on the historical structure sequence and historical risk time of each test coal mine support, construct a time prediction sample, and train a risk state prediction model based on the time prediction sample;

[0120] Step 4: Collect the initial environment sequence of the coal mine support to be monitored and the initial structure sequence of each structure parameter;

[0121] Step 5: Based on the initial environment sequence and the initial structure sequence, a structure prediction instance is constructed, and based on the structure prediction instance and the future value prediction model, the future prediction values ​​of each structure parameter are obtained;

[0122] Step 6: Based on the future predicted values ​​of each structural parameter, a time prediction instance is constructed, and based on the time prediction instance and the risk status prediction model, a safety risk warning is initiated;

[0123] Step 7: Update the initial environment sequence and the initial structure sequence over time, and execute steps 5 to 6 repeatedly.

[0124] Example 3

[0125] 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, according to another aspect of the present application, an electronic device 100 is also provided. The electronic device 100 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, the above-mentioned coal mine support safety risk prediction and early warning method based on big data analysis can be executed.

[0126] The method or device 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 100 may include a bus 101, one or more CPUs 102, ROM 103, RAM 104, a communication port 105 connected to a network, an input / output component 106, a hard disk 107, etc. The storage device in the electronic device 100, such as ROM 103 or hard disk 107, may store the coal mine support safety risk prediction and early warning method based on big data analysis provided in this application.

[0127] Furthermore, the electronic device 100 may further include a user interface 108. 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.

[0128] Example 4

[0129] Figure 4 Schematic diagram of the computer-readable storage medium structure provided by an embodiment of the present application. Figure 4As shown, a computer-readable storage medium 200 according to an embodiment of the present application is shown. Computer-readable instructions are stored on the computer-readable storage medium 200. When the computer-readable instructions are executed by the processor, the method for predicting and warning the safety risks of coal mine support based on big data analysis according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium 200 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), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0130] In addition, according to the implementation of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, the non-transitory machine-readable storage medium stores machine-readable instructions, the machine-readable instructions can be run by a processor to execute instructions corresponding to the method steps provided by the present application, and when the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0131] The methods, apparatuses, and devices of the present application may be implemented in many ways. For example, the methods, apparatuses, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, 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 may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the method according to the present application. Therefore, the present application also covers recording media storing programs for executing the method according to the present application.

[0132] 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.

[0133] 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.

[0134] The above preset parameters or preset thresholds are all set by technicians in this field according to actual conditions or obtained through large-scale data simulation.

[0135] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A coal mine support safety risk prediction and early warning system based on big data analysis, characterized in that: It includes a historical data collection module, a structural model training module, a risk model training module and a risk monitoring module; wherein each module is electrically connected; The historical data collection module collects in advance the historical environmental sequences of various environmental parameters of each test coal mine support during the test process, the historical structural sequences of various structural parameters changing over time, and the historical risk time of safety risks, and sends the historical environmental sequences and historical structural sequences to the structural model training module; and sends the historical environmental sequences, historical structural sequences and historical risk time to the risk model training module; The structural model training module constructs structural prediction samples for various structural parameters based on the historical environmental sequences and historical structural sequences of each test coal mine support, trains future value prediction models for various structural parameters based on the structural prediction samples, and sends the trained future value prediction models to the risk monitoring module; The risk model training module constructs time prediction samples based on the historical structure sequence and historical risk time of each test coal mine support, trains the risk status prediction model based on the time prediction samples, and sends the trained risk status prediction model to the risk monitoring module; The risk monitoring module collects the initial environment sequence of the coal mine support to be monitored and the initial structure sequence of each structural parameter, constructs a structure prediction instance based on the initial environment sequence and the initial structure sequence, obtains the future prediction value of each structural parameter based on the structure prediction instance and the future value prediction model, constructs a time prediction instance based on the future prediction value of each structural parameter, and initiates a safety risk warning based on the time prediction instance and the risk status prediction model; and updates the initial environment sequence and the initial structure sequence over time, cyclically constructing structure prediction instances, time prediction instances and safety risk warnings.

2. The coal mine support safety risk prediction and early warning system based on big data analysis according to claim 1 is characterized in that: The collection of the historical environment sequence, the historical structure sequence of changes of various structural parameters over time, and the historical risk time includes the following steps: Step 11: Arrange environmental parameter sensors for various environmental parameters in the area where the test coal mine support installation position is located, and collect parameter values ​​of various environmental parameters in real time through the environmental parameter sensors; the parameter sequence of various environmental parameters arranged in chronological order is regarded as a set of historical environmental sequences; Step 12: Install structural parameter sensors corresponding to various structural parameters at key support positions of the test coal mine support, and collect parameter values ​​of various structural parameters in real time through various structural parameter sensors; the key support positions are manually selected in advance according to the physical structural characteristics of the coal mine support structure; the structural sequence of various structural parameters arranged in chronological order is taken as a set of historical structural sequences; Step 13: Coal mine site personnel record the time when safety problems occur in each coal mine support structure as the historical risk time.

3. The coal mine support safety risk prediction and early warning system based on big data analysis according to claim 2 is characterized in that: The method of constructing structural prediction samples for various structural parameters based on the historical environment sequence and historical structure sequence of each test coal mine support is as follows: Preset prediction time length t; Preset the time window length w and sliding step size s; For each structural parameter of each tested coal mine support: With a time window length w and a sliding step length s, a sliding window method is used to obtain environmental subsequences and structural subsequences with a length of the time window length w from each historical environmental sequence and the historical structural sequence of the corresponding structural parameters. The environmental subsequences and structural subsequences obtained at each sliding constitute the structural feature samples. The parameter value of the structural parameter corresponding to the moment of prediction time length t after each sliding moment in the historical structural sequence is used as the structural label sample; Each group of structural feature samples and structural label samples constitutes a structural prediction sample.

4. The coal mine support safety risk prediction and early warning system based on big data analysis according to claim 3 is characterized in that: The method of training the future value prediction model for each structural parameter based on the structural prediction sample is as follows: For each structural parameter: The structural feature sample of each structural prediction sample is used as the input of the future value prediction model, and the future value prediction model takes the predicted value of the structural parameter after the prediction time length t as the output; the future value prediction model takes the structural label sample corresponding to the structural prediction sample as the prediction target, and takes the difference between the predicted value of the structural parameter and the structural label sample as the first prediction error, and takes minimizing the sum of squares of the first prediction errors of all structural prediction samples as the training target; the future value prediction model is trained until the sum of squares of the first prediction error reaches convergence and the training is stopped; the future value prediction model is any one of the time series prediction models.

5. The coal mine support safety risk prediction and early warning system based on big data analysis according to claim 4 is characterized in that: The method of constructing the time prediction sample based on the historical structure sequence and historical risk time of each test coal mine support is as follows: For each tested coal support: The parameter values ​​of each structural parameter at each moment are combined into a set of time prediction vectors; A risk status label is set for each moment; the risk status label is set in such a way that if the moment is a historical risk time when a security risk occurs, the risk status label is set to 1; if the moment is not a historical risk time when a security risk occurs, the risk status label is set to 0; The time prediction vector and risk status label at each moment constitute the time prediction sample.

6. The coal mine support safety risk prediction and early warning system based on big data analysis according to claim 5 is characterized in that: The method of training the risk status prediction model based on the time prediction sample is as follows: The time prediction vector of each test coal mine support at each moment is used as the input of the risk state prediction model, and the risk state prediction model takes the predicted value of the risk state at each moment as the output, and the predicted value of the risk state is 0 or 1; the risk state prediction model takes the risk state label corresponding to each moment as the prediction target, takes the difference between the predicted value of the risk state and the risk state label as the second prediction error, and takes minimizing the sum of squares of the second prediction errors at all moments as the training target; The risk status prediction model is trained until the sum of squares of the second prediction errors reaches convergence, and the training is stopped; the risk status prediction model is any one of the classification models.

7. The coal mine support safety risk prediction and early warning system based on big data analysis according to claim 6 is characterized in that: The method of obtaining the future prediction value of each structural parameter based on the structural prediction instance and the future value prediction model is as follows: For each structural parameter, the structural prediction instance is input into the future value prediction model, and the predicted value of the structural parameter after the prediction time length t output by the future value prediction model is obtained as the future prediction value; The method of constructing a time prediction instance based on the future prediction values ​​of each structural parameter and initiating a safety risk warning based on the time prediction instance and the risk status prediction model is as follows: The future prediction values ​​of each structural parameter are combined into a set of structural prediction value vectors as a set of time prediction instances; Input the structure prediction value vector into the risk state prediction model to obtain the predicted value of the risk state after the prediction time length t output by the risk state prediction model. If the predicted value of the risk state is 1, a safety risk alarm is initiated; Otherwise, no security risk alarm is initiated.

8. A method for predicting and warning safety risks of coal mine support based on big data analysis, which is implemented based on the system for predicting and warning safety risks of coal mine support based on big data analysis as described in any one of claims 1 to 7, characterized in that: The following steps are involved: Step 1: Collect in advance the historical environmental sequences of various environmental parameters of each test coal mine support during the test process, the historical structural sequences of various structural parameters changing over time, and the historical risk time of safety risks; Step 2: Based on the historical environment sequence and historical structure sequence of each test coal mine support, construct a structural prediction sample for each structural parameter, and based on the structural prediction sample, train a future value prediction model for each structural parameter; Step 3: Based on the historical structure sequence and historical risk time of each test coal mine support, construct a time prediction sample, and train a risk state prediction model based on the time prediction sample; Step 4: Collect the initial environment sequence of the coal mine support to be monitored and the initial structure sequence of each structure parameter; Step 5: Based on the initial environment sequence and the initial structure sequence, a structure prediction instance is constructed, and based on the structure prediction instance and the future value prediction model, the future prediction values ​​of each structure parameter are obtained; Step 6: Based on the future predicted values ​​of each structural parameter, a time prediction instance is constructed, and based on the time prediction instance and the risk status prediction model, a safety risk warning is initiated; Step 7: Update the initial environment sequence and the initial structure sequence over time, and execute steps 5 to 6 repeatedly.

9. An electronic device, characterized in that: include: processor and memory, wherein The memory stores a computer program that can be called by the processor; The processor executes the coal mine support safety risk prediction and early warning method based on big data analysis as described in claim 8 in the background by calling the computer program stored in the memory.

10. A computer-readable storage medium, characterized in that: A rewritable computer program is stored thereon; When the computer program runs on a computer device, the computer device executes the coal mine support safety risk prediction and early warning method based on big data analysis as described in claim 8 in the background.

Citation Information

Patent Citations

  • Forecasting and forewarning method and device for shale gas fracture downhole accidents

    CN107642347A

  • Vehicle window fogging prediction method and system, electronic equipment and storage medium

    CN112801202A

  • Historical building structure response prediction method and system in atmospheric temperature field

    CN115906672A

  • Intelligent monitoring and early warning system for historical building structure under stable load

    CN116090347A

  • Regional flood disaster grade prediction and early warning method and system

    CN116434479A

Cited By

  • Fault diagnosis method and system for valve actuating mechanism of steam turbine

    CN119756842A

  • Turbine valve actuator fault diagnosis method and system

    CN119756842B

  • Factory monitoring and safety system and method integrating multiple systems

    CN120065938A

  • A converged multi-system plant monitoring and safety system and method

    CN120065938B