Intelligent monitoring and control system for the advanced support traction in coal mining working face tunnel

Abstract

The present invention discloses an intelligent monitoring and control system for the traction of advance supports in coal mining working face tunnels, which relates to the field of intelligent monitoring technology. Sensors are used to continuously monitor the support stress and geological pressure fluctuations, and based on the extracted feature data, a machine learning model is used to predict the adaptability value of the control strategy to the geological conditions. The adaptability values ​​are divided into categories according to the adaptability values. The high adaptability category maintains the current operation, the low adaptability category stops the operation and issues an early warning, and the medium adaptability category optimizes the timing, speed and supporting force of the support movement through dynamic regulation, thereby achieving flexible adjustment of the control strategy under complex working conditions, and ultimately improving the system's adaptability to changes in geological conditions, preventing support failure and equipment chain damage, and improving the safety and efficiency of coal mining operations.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring technology, and in particular to an intelligent monitoring and control system for traction of advance supports in a coal mining working face tunnel. Background Art

[0002] Intelligent monitoring and control of the movement of the advance support in the coal mining face tunnel refers to the use of advanced sensors, automation equipment and intelligent control systems to monitor and accurately control the movement of the advance support in the coal mining face tunnel in real time. By collecting and analyzing the dynamic data of the support status, position and working environment, the support can be automatically adjusted and moved in the tunnel, ensuring the safety and efficiency of the coal mining process, reducing manual intervention and improving the stability and production efficiency of the working face.

[0003] In the prior art, when the advanced support in the coal mining face tunnel is intelligently monitored and controlled, first, the state, position and surrounding environment of the support are monitored in real time by sensors to collect data; then, the data is transmitted to the central control system for analysis and judgment, and the movement mode and timing of the support are determined according to the preset control strategy; then, the control system issues instructions to automatically control the hydraulic support to be moved or adjusted; finally, the position and state of the support are monitored again through the feedback system to ensure that the movement is in place and meets safety requirements.

[0004] The prior art has the following deficiencies:

[0005] The preset control strategy is usually designed based on typical working conditions, but the geological conditions and environmental factors in coal mining vary widely and are difficult to fully predict. If the control strategy is too rigid or fails to adjust dynamically according to the actual situation, it may lead to improper timing of support movement. At the same time, if the support moves improperly, is overstressed or moves at inappropriate times, it may cause damage to the support itself. After the support fails, it cannot provide effective support, which will cause surrounding equipment such as conveyors and tunneling equipment to be buried and damaged, causing chain damage to the equipment. Summary of the invention

[0006] The object of the present invention is to provide an intelligent monitoring and control system for the traction of advance supports in coal mining working face tunnels, so as to solve the shortcomings of the background technology.

[0007] In order to achieve the above-mentioned purpose, the present invention provides the following technical solution: an intelligent monitoring and control system for the traction of the advance support in the coal mining working face tunnel, including a monitoring module, a feature extraction module, a machine learning model training module, an adaptability evaluation module, a control strategy execution module and a dynamic control module:

[0008] Monitoring module: Several different types of sensors are arranged around the tunnels and supports of the coal mining face. The sensors are used to continuously monitor the support stress and geological pressure fluctuations and generate real-time data streams. Through the data acquisition system, the force data of the support and the environmental data of the tunnel are transmitted to the central control system in real time;

[0009] Feature extraction module: pre-process the acquired data, perform feature extraction on the pre-processed support stress data and tunnel environmental data, and respectively extract the response time characteristics of the support stress in the support stress data and the change characteristics of the geological pressure fluctuation in the tunnel environmental data;

[0010] Machine learning model training module: Based on the extracted response time characteristics of the support stress and the change characteristics of the geological pressure fluctuation, a first machine learning model for predicting the adaptability of the control strategy to the change of geological conditions is trained, and the adaptability value of the control strategy to the change of geological conditions is determined according to the output results of the model;

[0011] Adaptability evaluation module: compare and analyze the adaptability of the control strategy to the change of geological conditions with the gradient standard threshold, and divide the adaptability of the control strategy to the change of geological conditions into high adaptability, medium adaptability and low adaptability according to the comparison results;

[0012] Control strategy execution module: For high adaptability categories, continue to perform bracket movement operations according to the current control strategy; for low adaptability categories, immediately stop automatic control operations and issue an early warning to the operator;

[0013] Dynamic control module: For the medium adaptability category, the adaptability abnormality degree of the control strategy in the subsequent fixed time period is predicted and analyzed. If the abnormality degree is high, the control strategy is optimized and the timing, movement speed and support force of the bracket are dynamically controlled.

[0014] Preferably, in the feature extraction module, a stent stress response time anomaly index is generated according to the extracted stent stress response time feature, and the method for obtaining the stent stress response time anomaly index is:

[0015] Get the first bracket stress response time value As an initial value, determine the smoothing factor : Calculate the exponentially weighted moving average at the current moment using the exponentially weighted moving average formula , the expression is: ; In the formula, is the support stress response time at time t, for The exponentially weighted moving average of the time step, for each time step t, is calculated based on the new support stress response time value and The exponentially weighted moving average of the current moment is calculated ; Set the current support stress response time and By comparison, the abnormal index of the support stress response time is calculated, that is, the absolute value of the difference between the two, and the expression is: ; In the formula, is the abnormal index of stent stress response time.

[0016] Preferably, the tunnel geological pressure fluctuation index is generated according to the change characteristics of the geological pressure fluctuation in the extracted tunnel environment data, and the method for obtaining the tunnel geological pressure fluctuation index is:

[0017] The geological pressure sensor installed in the tunnel continuously collects the time series data P(s) of the geological pressure, where s represents the acquisition time and P(s) represents the geological pressure value at time s. ; Where N is the number of collected data points and the length of the time series; remove the trend term in the pressure data, the expression is: In the formula, For the pressure data after removing the trend, is the mean of the geological pressure data; the pressure data is windowed using the window function: ;W(s) is the window function, The pressure data after window function processing is transformed into time series using fast Fourier transform. Converted to frequency domain, the expression is: ; In the formula, For frequency The spectrum value at, that is, the frequency component, is the frequency component, k=0,1,…,N−1, i is the imaginary unit, is the kernel function of Fourier transform, which calculates the amplitude of the frequency component , the expression is: ; After analyzing the amplitudes of the calculated m main frequency components, the tunnel geological pressure fluctuation index is calculated, and the expression is: ; where k is the number of frequencies, It is the tunnel geological pressure fluctuation index.

[0018] Preferably, in the machine learning model training module, the support stress response time anomaly index and the tunnel geological pressure fluctuation index are converted into a first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses each group of first eigenvectors to predict the adaptability value label of the control strategy to the change in geological conditions as the prediction target, and minimizes the sum of the prediction errors of the adaptability value labels of all control strategies to the change in geological conditions as the training target. The machine learning model is trained until the sum of the prediction errors converges, then the model training is stopped, and the adaptability value of the control strategy to the change in geological conditions is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0019] Preferably, in the adaptability evaluation module, the adaptability value of the control strategy to the change of geological conditions is compared and analyzed with the gradient standard threshold, specifically:

[0020] Compare the acquired adaptability value of the control strategy to the geological condition change with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and compare the adaptability value of the control strategy to the geological condition change with the first standard threshold and the second standard threshold respectively;

[0021] If the adaptability value of the control strategy to the change of geological conditions is greater than the second standard threshold, it means that the adaptability of the control strategy to the change of geological conditions is high, and a high adaptability signal is generated at this time, and it is classified into a high adaptability category;

[0022] If the adaptability value of the control strategy to the geological condition change is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the adaptability of the control strategy to the geological condition change is average, and a medium adaptability signal is generated at this time, and it is classified as a medium adaptability category;

[0023] If the adaptability value of the control strategy to the change of geological conditions is less than the first standard threshold, it means that the adaptability of the control strategy to the change of geological conditions is low. At this time, a low adaptability signal is generated and it is classified into a low adaptability category.

[0024] Preferably, in the dynamic control module, for the medium adaptability category, the adaptability abnormality degree of the control strategy in the subsequent fixed time period is predicted and analyzed. If the abnormality degree is high, the control strategy is optimized, and the timing, movement speed and support force of the support are dynamically controlled;

[0025] Determine a fixed time period Δh, monitor and predict the adaptability of the control strategy during this time period, and predict the adaptability value :Use the control strategy feature vector and historical trend of the current time h to predict the future time through the polynomial regression model The control strategy adaptability value when ; Calculate the degree of abnormality , that is, calculating the difference between the actual fitness value and the predicted value, the expression is: ;in, It is a future time point The actual measured fitness value, is the predicted fitness value.

[0026] Preferably, the acquired abnormality degree value is compared with the abnormality degree threshold. If the abnormality degree is greater than or equal to the preset abnormality degree threshold, the adaptive abnormality degree is high and the control strategy needs to be optimized; if the abnormality degree is less than the preset abnormality degree threshold, the adaptive abnormality degree is low and the control strategy does not need to be optimized.

[0027] Preferably, if the abnormality is high, the control strategy is optimized to dynamically adjust the timing, movement speed and support force of the support, specifically:

[0028] Optimize the timing of bracket movement: set the adjustment parameters of the movement timing If the adaptability is abnormal, adjust the timing of the bracket movement , the expression is: ;in, is the adjustment amount, a positive value indicates a delayed move, a negative value indicates an early move. The timing of the adjusted bracket movement;

[0029] Bracket moving speed optimization: Set the moving speed adjustment parameter ΔV to adjust the bracket moving speed , the expression is: ; In the formula, ΔV is the adjustment amount, which increases or decreases the support moving speed according to the changes in geological conditions. is the adjusted bracket moving speed;

[0030] Bracket support force optimization: setting support force adjustment parameters , according to the predicted geological pressure fluctuation, adjust the support force of the support, the expression is: ;in, Dynamic adjustment of the support force to increase or decrease the support force of the bracket to ensure the stability of the bracket under abnormal conditions. is the support force of the bracket after adjustment.

[0031] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0032] 1. The present invention continuously monitors the support stress and geological pressure fluctuations by arranging sensors around the coal mining face tunnel and the support, and extracts the key stress response time characteristics and pressure fluctuation characteristics based on the feature extraction module, and uses the machine learning model to predict the adaptability of the control strategy to the changes in geological conditions. The adaptability evaluation module divides the control strategy into high adaptability, medium adaptability and low adaptability categories by comparing the adaptability value with the gradient standard threshold, thereby optimizing the support movement timing, speed and support force, and ensuring the flexibility of the system in dealing with complex geological conditions.

[0033] 2. The present invention maintains the current operation for high adaptability categories; immediately stops automatic operation and issues an early warning for low adaptability categories; and predicts and analyzes the degree of adaptability abnormality within a fixed time period for medium adaptability categories. If the degree of abnormality is high, the control strategy is dynamically adjusted, effectively solving the problems of rigid control strategy and inappropriate movement timing in the prior art, avoiding support damage and equipment chain damage, and improving the safety and efficiency of coal mining operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0035] Figure 1 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0036] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not 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.

[0037] For examples, see Figure 1 As shown, the intelligent monitoring and control system for the advanced support traction in the coal mining working face tunnel described in this embodiment includes a monitoring module, a feature extraction module, a machine learning model training module, an adaptability evaluation module, a control strategy execution module and a dynamic control module:

[0038] Monitoring module: Several different types of sensors are arranged around the tunnels and supports of the coal mining face. The sensors are used to continuously monitor the support stress and geological pressure fluctuations and generate real-time data streams. Through the data acquisition system, the force data of the support and the environmental data of the tunnel are transmitted to the central control system in real time;

[0039] Feature extraction module: pre-process the acquired data, perform feature extraction on the pre-processed support stress data and tunnel environmental data, and respectively extract the response time characteristics of the support stress in the support stress data and the change characteristics of the geological pressure fluctuation in the tunnel environmental data;

[0040] Machine learning model training module: Based on the extracted response time characteristics of the support stress and the change characteristics of the geological pressure fluctuation, a first machine learning model for predicting the adaptability of the control strategy to the change of geological conditions is trained, and the adaptability value of the control strategy to the change of geological conditions is determined according to the output results of the model;

[0041] Adaptability evaluation module: compare and analyze the adaptability of the control strategy to the change of geological conditions with the gradient standard threshold, and divide the adaptability of the control strategy to the change of geological conditions into high adaptability, medium adaptability and low adaptability according to the comparison results;

[0042] Control strategy execution module: For high adaptability categories, continue to perform bracket movement operations according to the current control strategy; for low adaptability categories, immediately stop automatic control operations and issue an early warning to the operator;

[0043] Dynamic control module: For the medium adaptability category, the adaptability abnormality degree of the control strategy in the subsequent fixed time period is predicted and analyzed. If the abnormality degree is high, the control strategy is optimized and the timing, movement speed and support force of the bracket are dynamically controlled.

[0044] In the monitoring module, stress sensors are installed on the key stress-bearing parts of the support, such as the support top plate, bottom plate and supporting hydraulic cylinder, to continuously monitor the pressure on the support, especially the stress of the top plate and bottom plate. Geological pressure sensors are installed around the tunnel wall, top plate and support to monitor the pressure changes from the strata, especially to capture the fluctuation of mine pressure in real time during the advancement of the working face. Displacement sensors are installed between the support top plate and the tunnel top plate to monitor the displacement or sinking of the tunnel top plate, which is used to evaluate the support effect and tunnel stability. Temperature sensors, humidity sensors, gas concentration sensors, etc. are arranged around the tunnel and the support to monitor the temperature, humidity and gas concentration of the tunnel environment in real time to ensure the safety of the coal mining environment.

[0045] Each sensor continuously collects the stress conditions of the support, geological pressure fluctuations, roof displacement and environmental data, and generates a real-time data stream. The data includes: support stress data, geological pressure fluctuation data, roof displacement data, tunnel environment data (temperature, humidity, gas concentration, etc.). Each sensor is connected to the data acquisition system wirelessly (Wi-Fi, Zigbee, etc.) or wired (Ethernet, optical fiber, etc.). The system will collect real-time data from each sensor every second or more. The data stream is transmitted to the central control system through a high-speed transmission network (such as industrial Ethernet or 5G network). In order to avoid data loss or transmission delay, the system supports a redundant transmission mechanism, that is, key data is transmitted simultaneously through multiple transmission channels to ensure the stability and real-time performance of the transmission.

[0046] In the feature extraction module, after the central control system receives the real-time transmitted data, it first stores the data in a distributed database to ensure efficient access and historical data backtracking capabilities. The force data of the support and the environmental data of the tunnel are stored in the corresponding database tables respectively. After obtaining the force data of the support and the environmental data of the tunnel, preprocessing operations are first performed to ensure the accuracy, completeness and consistency of the data. Including: using filtering techniques (such as Kalman filtering or low-pass filtering) to eliminate random noise in the data to ensure smoother and more accurate measurements. Detecting extreme outliers and processing them by interpolation or elimination. For example, using the 3σ criterion or boxplot analysis to identify and process data points that are significantly deviated from the normal range to avoid these outliers affecting subsequent analysis. Since data from different types of sensors may have different acquisition frequencies, time synchronization processing is required. Due to sensor failure or environmental factors, some data may be lost. The system repairs missing data through interpolation or regression filling. In order to make different sensor data have consistent metrics, normalization or standardization processing is used. For example, for stress data and pressure data, Min-Max normalization can be used to scale the data to the range of [0, 1] so that subsequent machine learning algorithms can more effectively process data with different characteristics.

[0047] After completing data preprocessing, the system extracts key features from the support force data and tunnel environment data for subsequent model training and analysis. Two main features are extracted: the response time characteristics of the support stress and the change characteristics of geological pressure fluctuations. The support stress response time anomaly index is generated based on the extracted support stress response time characteristics. The method for obtaining the support stress response time anomaly index is as follows:

[0048] Get the first bracket stress response time value As an initial value, determine the smoothing factor : α is used as the weight decay factor in the exponentially weighted moving average, which determines the weight of the most recent response time value in the total value. The value range of α is 0<α≤1: if α is large (such as close to 1), the most recent data points have a greater impact and the reaction is faster; if α is small, the historical data has a greater impact and the reaction is slower. Usually, α is set between 0.1 and 0.3, and the specific value can be adjusted according to the system's sensitivity requirements to anomalies. For example, α=0.1 (example value); the exponentially weighted moving average at the current moment is calculated using the exponentially weighted moving average formula , the expression is: ; In the formula, is the support stress response time at time t, for The exponentially weighted moving average of the time step, for each time step t, is calculated based on the new support stress response time value and The exponentially weighted moving average of the current moment is calculated ; Set the current support stress response time and Calculate the abnormal index of the bracket stress response time, that is, the absolute value of the difference between the two, and the expression is: ; In the formula, is the abnormal index of stent stress response time.

[0049] The larger the abnormal index of the support stress response time, the greater the deviation of the current support stress response time from the historical trend, indicating that the control strategy is less adaptable to changes in geological conditions. This means that the support cannot make timely or accurate adjustments when facing geological pressure fluctuations, resulting in delayed response time or exceeding the normal range, which may increase the risk of support failure and tunnel instability. Therefore, a larger abnormal index indicates that the control strategy needs to be optimized to improve the system's responsiveness and stability under dynamic geological conditions.

[0050] The tunnel geological pressure fluctuation index is generated according to the change characteristics of the geological pressure fluctuation in the extracted tunnel environment data. The method for obtaining the tunnel geological pressure fluctuation index is as follows:

[0051] The geological pressure sensor installed in the tunnel continuously collects the time series data P(s) of the geological pressure, where s represents the acquisition time and P(s) represents the geological pressure value at time s. ; Where N is the number of collected data points and the length of the time series; remove the trend term in the pressure data, the expression is: ; In the formula, For the pressure data after removing the trend, is the mean of the geological pressure data; use a window function (such as a Hanning window or a Hamming window) to perform window processing on the pressure data: ;W(s) is the window function, commonly used ones include Hanning window, rectangular window, etc. The pressure data after window function processing is transformed into time series using fast Fourier transform. Converted to frequency domain, the expression is: ; In the formula, For frequency The spectrum value at, that is, the frequency component, is the frequency component, k=0,1,…,N−1, i is the imaginary unit, is the kernel function of Fourier transform, which calculates the amplitude of the frequency component , the expression is: ; After analyzing the amplitudes of the calculated m main frequency components, the tunnel geological pressure fluctuation index is calculated, and the expression is: ; where k is the number of frequencies, It is the tunnel geological pressure fluctuation index.

[0052] The larger the tunnel geological pressure fluctuation index is, the more drastic the fluctuation amplitude and frequency of the geological pressure in the tunnel is, reflecting the increasing instability of the geological conditions. At this time, the control strategy is less adaptable to changes in geological conditions, because larger fluctuations may cause the support and tunnel support system to be unable to respond to sudden pressure changes in a timely or effective manner, increasing the risk of support failure or tunnel collapse. Therefore, a larger fluctuation index indicates that the control strategy needs to be adjusted or optimized to improve the system's ability to respond to complex and drastic geological changes.

[0053] Machine learning model training module: Based on the extracted response time characteristics of the support stress and the changing characteristics of the geological pressure fluctuations, the first machine learning model for predicting the adaptability of the control strategy to changes in geological conditions is trained, and the adaptability value of the control strategy to changes in geological conditions is determined based on the model output results.

[0054] The support stress response time anomaly index and the tunnel geological pressure fluctuation index are converted into the first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses the adaptability value label of each group of first eigenvectors to predict the change of geological conditions as the prediction target, and minimizes the sum of the prediction errors of the adaptability value labels of all control strategies to the change of geological conditions as the training target. The machine learning model is trained until the sum of the prediction errors converges, then the model training is stopped, and the adaptability value of the control strategy to the change of geological conditions is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0055] The method for obtaining the adaptability value of the control strategy to the change of geological conditions is: from the first eigenvector training data of the trained machine learning model, obtain the corresponding function expression: ; In the formula, is the output function of the model, is the abnormal index of the stent stress response time, is the tunnel geological pressure fluctuation index, is the adaptability of the control strategy to changes in geological conditions.

[0056] Adaptability evaluation module: Compare and analyze the adaptability value of the control strategy to changes in geological conditions with the gradient standard threshold, and divide the adaptability of the control strategy to changes in geological conditions into high adaptability category, medium adaptability category and low adaptability category according to the comparison results.

[0057] Compare the acquired adaptability value of the control strategy to the geological condition change with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and compare the adaptability value of the control strategy to the geological condition change with the first standard threshold and the second standard threshold respectively;

[0058] If the adaptability value of the control strategy to the change of geological conditions is greater than the second standard threshold, it means that the adaptability of the control strategy to the change of geological conditions is high, and a high adaptability signal is generated at this time, and it is classified into a high adaptability category;

[0059] If the adaptability value of the control strategy to the geological condition change is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the adaptability of the control strategy to the geological condition change is average, and a medium adaptability signal is generated at this time, and it is classified as a medium adaptability category;

[0060] If the adaptability value of the control strategy to the change of geological conditions is less than the first standard threshold, it means that the adaptability of the control strategy to the change of geological conditions is low. At this time, a low adaptability signal is generated and it is classified into a low adaptability category.

[0061] Control strategy execution module: For high adaptability categories, continue to perform bracket movement operations according to the current control strategy; for low adaptability categories, immediately stop automatic control operations and issue an early warning to the operator.

[0062] If the adaptability value of the control strategy to changes in geological conditions is greater than the second standard threshold, it means that the control strategy is highly adaptable to changes in geological conditions, indicating that the current geological pressure fluctuations are within the controllable range, and the support and control system can effectively respond to changes in geological pressure. In this case, continue to move the support according to the current control strategy without adjusting the support's movement method, support force, or movement timing. This means that the system currently has sufficient adaptability to changes in geological pressure and can ensure support stability and roadway safety.

[0063] If the adaptability value of the control strategy to changes in geological conditions is less than the first standard threshold, it means that the control strategy has a low degree of adaptability to changes in geological conditions, the tunnel structure may be unstable, and the current control strategy is difficult to effectively respond to geological changes. There is a high risk of support failure and tunnel collapse. The system immediately stops the automatic movement and adjustment of the support to prevent movement operations when the pressure fluctuates violently, so as to avoid the support from becoming unstable or moving improperly, which will increase the tunnel pressure. The system sends an early warning signal to the operator, reminding the operator that the current geological conditions are abnormal and requiring manual intervention for inspection and adjustment. The operator can manually control the movement and support of the support to ensure safety.

[0064] Dynamic control module: For the medium adaptability category, the adaptability abnormality degree of the control strategy in the subsequent fixed time period is predicted and analyzed. If the abnormality degree is high, the control strategy is optimized and the timing, movement speed and support force of the bracket are dynamically controlled.

[0065] Determine a fixed time period Δh, monitor and predict the adaptability of the control strategy during this time period, and predict the adaptability value :Use the control strategy feature vector and historical trend of the current time h to predict the future time through the polynomial regression model The control strategy adaptability value when ;

[0066] Calculate the degree of abnormality , that is, calculating the difference between the actual fitness value and the predicted value, the expression is: ;in, It is a future time point The actual measured fitness value, is the predicted fitness value.

[0067] The acquired abnormality degree value is compared with the abnormality degree threshold. If the abnormality degree is greater than or equal to the preset abnormality degree threshold, it is considered that the adaptive abnormality degree is high and the control strategy needs to be optimized; if the abnormality degree is less than the preset abnormality degree threshold, it is considered that the adaptive abnormality degree is low and the control strategy does not need to be optimized.

[0068] If the abnormality is high, the control strategy is optimized to dynamically adjust the timing, speed and support force of the bracket movement, specifically:

[0069] Optimize the timing of bracket movement: set the adjustment parameters of the movement timing If the adaptability is abnormal, adjust the timing of the bracket movement , the expression is: ;in, is the adjustment amount, a positive value indicates a delayed move, a negative value indicates an early move. The timing of the adjusted bracket movement;

[0070] Bracket moving speed optimization: Set the moving speed adjustment parameter ΔV to adjust the bracket moving speed , the expression is: ; In the formula, ΔV is the adjustment amount, which increases or decreases the support moving speed according to the changes in geological conditions. is the adjusted bracket moving speed;

[0071] Bracket support force optimization: setting support force adjustment parameters , according to the predicted geological pressure fluctuation, adjust the support force of the support, the expression is: ;in, Dynamic adjustment of the support force to increase or decrease the support force of the bracket to ensure the stability of the bracket under abnormal conditions. is the support force of the bracket after adjustment.

[0072] It should be noted here that , ΔV, The specific value of is obtained by analyzing historical data, including geological pressure fluctuations, support movement timing, speed, support force and other parameters, and establishing a relationship model between these variables and geological pressure conditions. The relationship between them can be found through multiple linear regression or polynomial regression, which will not be described in detail here.

[0073] In this embodiment, several sensors are arranged around the tunnels and supports of the coal mining face, and the support stress and geological pressure fluctuations are continuously monitored by the monitoring module, and the data is transmitted to the central control system in real time. The feature extraction module preprocesses the collected data, and extracts the response time characteristics of the support stress and the change characteristics of the geological pressure fluctuations in the tunnel respectively. Based on these characteristics, the machine learning model training module trains a model that predicts the adaptability of the control strategy to the changes in geological conditions and generates an adaptability value. The adaptability evaluation module divides the control strategy into high, medium and low adaptability categories by comparing the adaptability value with the gradient standard threshold. The control strategy execution module operates according to the category: for the high adaptability category, the current support operation continues; for the low adaptability category, the automatic operation is stopped and an early warning is issued. For the medium adaptability category, the dynamic control module predicts the degree of adaptability abnormality. If the degree of abnormality is high, the timing, speed and support force of the support movement are optimized to ensure the safety and adaptability of the system.

[0074] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. An intelligent monitoring and control system for the forward support traction in the coal mining working face tunnel, characterized by: It includes monitoring module, feature extraction module, machine learning model training module, adaptability evaluation module, control strategy execution module and dynamic regulation module: Monitoring module: Several different types of sensors are arranged around the tunnels and supports of the coal mining face. The sensors are used to continuously monitor the support stress and geological pressure fluctuations and generate real-time data streams. Through the data acquisition system, the force data of the support and the environmental data of the tunnel are transmitted to the central control system in real time; Feature extraction module: pre-process the acquired data, perform feature extraction on the pre-processed support stress data and tunnel environmental data, and respectively extract the response time characteristics of the support stress in the support stress data and the change characteristics of the geological pressure fluctuation in the tunnel environmental data; Machine learning model training module: Based on the extracted response time characteristics of the support stress and the change characteristics of the geological pressure fluctuation, a first machine learning model for predicting the adaptability of the control strategy to the change of geological conditions is trained, and the adaptability value of the control strategy to the change of geological conditions is determined according to the output results of the model; Adaptability evaluation module: compare and analyze the adaptability of the control strategy to the change of geological conditions with the gradient standard threshold, and divide the adaptability of the control strategy to the change of geological conditions into high adaptability, medium adaptability and low adaptability according to the comparison results; Control strategy execution module: For high adaptability categories, continue to perform bracket movement operations according to the current control strategy; for low adaptability categories, immediately stop automatic control operations and issue an early warning to the operator; Dynamic control module: For the medium adaptability category, the adaptability abnormality degree of the control strategy in the subsequent fixed time period is predicted and analyzed. If the abnormality degree is high, the control strategy is optimized and the timing, movement speed and support force of the bracket are dynamically controlled.

2. The intelligent monitoring and control system for the advanced support traction in the coal mining working face tunnel according to claim 1 is characterized in that: In the feature extraction module, the stent stress response time anomaly index is generated according to the extracted stent stress response time feature. The method for obtaining the stent stress response time anomaly index is as follows: Get the first bracket stress response time value As an initial value, determine the smoothing factor : Calculate the exponentially weighted moving average at the current moment using the exponentially weighted moving average formula , the expression is: ; In the formula, is the support stress response time at time t, for The exponentially weighted moving average of the time step, for each time step t, is calculated based on the new support stress response time value and The exponentially weighted moving average of the current moment is calculated ; Set the current support stress response time and By comparison, the abnormal index of the support stress response time is calculated, that is, the absolute value of the difference between the two, and the expression is: ; In the formula, is the abnormal index of stent stress response time.

3. The intelligent monitoring and control system for the advanced support traction in the coal mining working face tunnel according to claim 2 is characterized in that: The tunnel geological pressure fluctuation index is generated according to the change characteristics of the geological pressure fluctuation in the extracted tunnel environment data. The method for obtaining the tunnel geological pressure fluctuation index is as follows: The geological pressure sensor installed in the tunnel continuously collects the time series data P(s) of the geological pressure, where s represents the acquisition time and P(s) represents the geological pressure value at time s. ; Where N is the number of collected data points and the length of the time series; remove the trend term in the pressure data, the expression is: ; In the formula, For the pressure data after removing the trend, is the mean of the geological pressure data; the pressure data is windowed using the window function: ;W(s) is the window function, The pressure data after window function processing is transformed into time series using fast Fourier transform. Converted to frequency domain, the expression is: ; In the formula, For frequency The spectrum value at, that is, the frequency component, is the frequency component, k=0,1,…,N−1, i is the imaginary unit, is the kernel function of Fourier transform, which calculates the amplitude of the frequency component , the expression is: ; After analyzing the amplitudes of the calculated m main frequency components, the tunnel geological pressure fluctuation index is calculated, and the expression is: ; where k is the number of frequencies, It is the tunnel geological pressure fluctuation index.

4. The intelligent monitoring and control system for the advanced support traction in the coal mining working face tunnel according to claim 3 is characterized in that: In the machine learning model training module, the support stress response time anomaly index and the tunnel geological pressure fluctuation index are converted into the first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses each group of first eigenvectors to predict the adaptability value label of the control strategy to the change of geological conditions as the prediction target, and minimizes the sum of the prediction errors of the adaptability value labels of all control strategies to the change of geological conditions as the training target. The machine learning model is trained until the sum of the prediction errors converges, and the model training is stopped. The adaptability value of the control strategy to the change of geological conditions is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

5. The intelligent monitoring and control system for the advanced support traction in the coal mining working face tunnel according to claim 4 is characterized in that: In the adaptability assessment module, the adaptability value of the control strategy to the change of geological conditions is compared and analyzed with the gradient standard threshold, specifically: Compare the acquired adaptability value of the control strategy to the geological condition change with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and compare the adaptability value of the control strategy to the geological condition change with the first standard threshold and the second standard threshold respectively; If the adaptability value of the control strategy to the change of geological conditions is greater than the second standard threshold, it means that the adaptability of the control strategy to the change of geological conditions is high, and a high adaptability signal is generated at this time, and it is classified into a high adaptability category; If the adaptability value of the control strategy to the geological condition change is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the adaptability of the control strategy to the geological condition change is average, and a medium adaptability signal is generated at this time, and it is classified as a medium adaptability category; If the adaptability value of the control strategy to the change of geological conditions is less than the first standard threshold, it means that the adaptability of the control strategy to the change of geological conditions is low. At this time, a low adaptability signal is generated and it is classified into a low adaptability category.

6. The intelligent monitoring and control system for the advanced support traction in the coal mining working face tunnel according to claim 1 is characterized in that: In the dynamic control module, for the medium adaptability category, the adaptability abnormality degree of the control strategy in the subsequent fixed time period is predicted and analyzed. If the abnormality degree is high, the control strategy is optimized, and the timing, movement speed and support force of the support are dynamically controlled; Determine a fixed time period Δh, monitor and predict the adaptability of the control strategy during this time period, and predict the adaptability value :Use the control strategy feature vector and historical trend of the current time h to predict the future time through the polynomial regression model The control strategy adaptability value when ; Calculate the degree of abnormality , that is, calculating the difference between the actual fitness value and the predicted value, the expression is: ;in, It is a future time point The actual measured fitness value, is the predicted fitness value.

7. The intelligent monitoring and control system for the advanced support traction in the coal mining face tunnel according to claim 6 is characterized in that: The acquired abnormality degree value is compared with the abnormality degree threshold. If the abnormality degree is greater than or equal to the preset abnormality degree threshold, the adaptive abnormality degree is high and the control strategy needs to be optimized; if the abnormality degree is less than the preset abnormality degree threshold, the adaptive abnormality degree is low and the control strategy does not need to be optimized.

8. The intelligent monitoring and control system for the advanced support traction in the coal mining face tunnel according to claim 7 is characterized in that: If the abnormality is high, the control strategy is optimized to dynamically adjust the timing, movement speed and support force of the bracket, specifically: Optimize the timing of bracket movement: set the adjustment parameters of the movement timing If the adaptability is abnormal, adjust the timing of the bracket movement , the expression is: ;in, is the adjustment amount, a positive value indicates a delayed move, a negative value indicates an early move. The timing of the adjusted bracket movement; Bracket moving speed optimization: Set the moving speed adjustment parameter ΔV to adjust the bracket moving speed , the expression is: ; In the formula, ΔV is the adjustment amount, which increases or decreases the support moving speed according to the changes in geological conditions. is the adjusted bracket moving speed; Bracket support force optimization: setting support force adjustment parameters , according to the predicted geological pressure fluctuation, adjust the support force of the support, the expression is: ;in, Dynamic adjustment of the support force to increase or decrease the support force of the bracket to ensure the stability of the bracket under abnormal conditions. is the support force of the bracket after adjustment.

Images