Mine tunnel operation period deformation monitoring method and system

Abstract

The invention discloses a mine tunnel operation period deformation monitoring method and system, and relates to the technical field of mine tunnel monitoring, and the method comprises the following steps: carrying out the real-time monitoring of the deformation condition in a mine tunnel through a sensor device, and transmitting the collected monitoring data to a data processing center through a wireless transmission module. Real-time monitoring of mine tunnel deformation is achieved through the multiple sensors and the wireless transmission technology, the problem of data lag is solved, and monitoring accuracy and timeliness are improved. And in combination with a multi-objective optimization algorithm, the system automatically adjusts key parameters such as a supporting structure and a mining sequence according to real-time data, so that an optimal safety control scheme is ensured. According to the system, the operation efficiency and safety are improved, manual misjudgment is reduced, and the control strategy can be continuously optimized according to the operation stage.

Description

Technical Field

[0001] The present invention relates to the technical field of mine tunnel monitoring, and in particular to a method and system for monitoring deformation of a mine tunnel during operation. Background Art

[0002] Deformation monitoring during the operation period of mine tunnels refers to the real-time monitoring and analysis of the deformation of tunnels and their surrounding geological environment by using various monitoring technologies during the operation stage of mine tunnels. These monitoring mainly include the measurement of deformation, displacement, stress, ground subsidence, etc. of tunnel structures to ensure the safe operation of tunnels and prevent or early detection of possible structural problems or safety hazards. Since mine tunnels are in a complex underground geological environment for a long time, factors such as geological changes, mining pressure, and rock sliding may cause tunnel deformation. Therefore, deformation monitoring is an important means to ensure tunnel stability, timely repair hidden dangers, and reduce accidents. Monitoring data can provide a scientific basis for engineering maintenance, design optimization, and safety decision-making, and ensure the continued safety of tunnel operations.

[0003] The prior art has the following deficiencies:

[0004] When applied to deformation monitoring during the operation period of mine tunnels, mine tunnels are usually in a complex underground environment. Factors such as geological pressure, rock movement, and mining activities may cause deformation at any time, especially sudden deformation in local areas. If there is a delay in the data collection and feedback process of the monitoring system, there will be a lag between the actual deformation and the monitoring results, and the key early warning window may be missed. This lag will cause the monitoring personnel to fail to identify the potential dangers of tunnel deformation in time, thereby delaying the handling of structural safety issues, increasing the risk of tunnel collapse, landslides or other disasters, and ultimately may lead to major accidents. Especially in the later stages of mine tunnel operation, underground pressure may gradually accumulate. If the real-time performance of the monitoring system is insufficient, it will greatly affect the implementation of safety warnings and risk management measures.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The purpose of the present invention is to provide a deformation monitoring method and system for mine tunnel operation. Through a variety of sensor devices and wireless transmission technology, real-time monitoring of mine tunnel deformation is achieved, solving the data lag problem in traditional monitoring methods, ensuring real-time collection and rapid feedback, greatly improving the accuracy and timeliness of monitoring, and enhancing tunnel safety. At the same time, the system adopts a multi-objective optimization algorithm (MOGA) to perform automated safety management and operation optimization based on real-time data, and adjusts key parameters (such as support structure, mining sequence, etc.) through intelligent algorithms to ensure the best safety control solution. This system not only improves operational efficiency and safety, but also reduces misjudgments caused by manual intervention. It can continuously optimize control strategies as the operation stage of the mine tunnel changes to solve the problems in the above-mentioned background technology.

[0007] In order to achieve the above object, the present invention provides the following technical solution: a deformation monitoring method applied to a mine tunnel during operation, comprising the following steps:

[0008] Use sensor equipment to monitor the deformation in the mine tunnel in real time, and transmit the collected monitoring data to the data processing center through the wireless transmission module;

[0009] Based on the historical data and field sensor data collected in real time, an adaptive weighted regression model is used to dynamically adjust the weight factor according to the historical data of different parts of the tunnel. The collected data is input into the regression model, and the deformation trend of the tunnel in the future is predicted using the linear fitting method, and the prediction results are updated in real time.

[0010] Combined with the deformation prediction results, the deep learning-based anomaly detection algorithm is used to analyze the monitoring data and identify potential sudden deformation risks. When the prediction results exceed the set safety threshold, the system triggers an early warning signal to remind operators to take emergency measures. At the same time, based on the deformation trend, the risk level and accident probability are evaluated to provide a decision-making basis for the safety management of mine tunnels.

[0011] With the support of deformation monitoring and early warning mechanism, combined with multi-objective optimization algorithm, the operation of mine tunnels is dynamically adjusted, and the parameters in the operation process are automatically adjusted according to the monitoring results to achieve the optimal safety control strategy, and the operation management of mine tunnels is optimized in real time under feedback control.

[0012] Preferably, the specific steps of using sensor equipment to monitor the deformation in the mine tunnel in real time and transmitting the collected monitoring data to the data processing center through the wireless transmission module are as follows;

[0013] In a mine tunnel, sensors must first be properly deployed based on geological conditions, tunnel structure, and deformation monitoring requirements;

[0014] When installing the sensor, ensure that the sensor can work stably and accurately in the tunnel environment;

[0015] Once the sensor is installed and debugged successfully, the system will start collecting data in real time;

[0016] After data collection, the sensor sends the data to the data processing center through the wireless transmission module.

[0017] Preferably, based on the historical data and field sensor data collected in real time, the deformation trend of the tunnel in the future is predicted by a linear fitting method, and the specific steps of updating the prediction results in real time are as follows:

[0018] First, the historical deformation data of the mine tunnel during operation is analyzed to provide a data basis for the regression model. The calculation expression is as follows:

[0019] , where D t is the deformation at time point t, X it is the measurement data of sensor i at time point t in the historical data, w i is the weight factor of sensor i at time point t, ∈ t is the error introduced in the model prediction, which represents the difference between the actual deformation and the model prediction, and n is the total number of sensors;

[0021] With the continuous input of real-time data, it is necessary to adjust the weight factors of each area according to the current monitoring data so that the model can respond more accurately to local sudden deformation. The adaptive weighted regression model is used to dynamically adjust the weight according to the monitored local deformation. The calculation expression is as follows:

[0022] w i (t) = w i (t-1)+η·δ i (t)

[0023] , where w i (t) represents the weight of sensor i at time point t, w i (t-1) is the weight of sensor i at time point t-1, η is the learning rate, which controls the amplitude of weight update, and δ i (t) is the weight change that sensor i needs to adjust at time point t, δ i The calculation formula of (t) is as follows:

[0024] , where ΔD i (t) is the change in deformation of sensor i at time point t, ΔD j (t) is the change in deformation of sensor j at time point t, It is the sum of the deformation changes measured by n sensors at time point t, reflecting the overall deformation trend of the mine tunnel at the current moment.

[0026] Preferably, after dynamically adjusting the weight factor, the sensor data collected in real time is input into the regression model for deformation prediction, and the calculation expression is as follows:

[0027] , where is the predicted value of tunnel deformation at time point t;

[0029] With the continuous increase of deformation monitoring data in the tunnel and the continuous optimization of the model, the prediction results also need to be updated in real time to further improve the prediction accuracy of the model. The sliding window method is used to roll the historical data. The calculation expression is as follows:

[0030] , where is the predicted deformation value after the time window T, w i (t-T+1) is the weight factor adjusted within the rolling window, X it-T+1 is the input data within the time window T.

[0032] Preferably, the steps of combining the deformation prediction results and applying the deep learning-based anomaly detection algorithm to identify potential sudden deformation risks are as follows:

[0033] First, the real-time monitoring data must be preprocessed and feature extracted;

[0034] Based on the preprocessed data, the anomaly detection model is trained using a deep learning algorithm;

[0035] After the deep learning model is deployed in the deformation monitoring system, the real-time monitoring data at each moment is automatically analyzed;

[0036] When the monitoring data is judged as abnormal by the model and exceeds the set safety threshold, further risk assessment is performed.

[0037] Preferably, with the support of deformation monitoring and early warning mechanism, combined with multi-objective optimization algorithm, the steps of dynamically adjusting the mine tunnel operation are as follows;

[0038] With the support of the deformation monitoring and early warning mechanism of mine tunnels, it is first necessary to integrate and summarize the real-time monitoring data obtained from the sensor network. The calculation expression is as follows:

[0039] F=λ1·f safety +λ2·f cost +λ3·f environment

[0040] , where F is the objective function, f safety represents the safety goal, f cost represents the economic benefit target, f environment represents the environmental impact target, λ1, λ2 and λ3 are the safety targets f safety 、Safety goal safety and economic benefit goals cost The weight coefficient of

[0041] In the multi-objective optimization process, in addition to the objective function, corresponding constraints need to be constructed to ensure that the decision in the optimization process is reasonable and meets the actual constraints. Assume that the stability force requirement of the support structure is R support , and with the support of monitoring data, ensure R support is greater than the threshold τ, the expression is as follows:

[0042] R support ≥τ

[0043] In addition, the mining sequence should be dynamically adjusted according to the real-time deformation data to ensure the optimal resource scheduling. The expression is as follows:

[0044] P mining ≤P safe (where P safe is a pre-defined safe mining rate)

[0045] , where P mining Indicates the current mining progress of the mine tunnel, P safe is a predefined safe mining rate;

[0046] When executing the multi-objective optimization algorithm, various objectives and constraints are comprehensively analyzed and optimized using methods such as simulated annealing, genetic algorithm or particle swarm optimization. Suppose the optimization result of the tunnel support structure is X support , the mining progress is adjusted to P mining , then the optimization problem can be transformed into the following form;

[0047] Maximize F(X support , P mining ,…)subject to C1, C2,…

[0048] , where F(X support , P mining , ...) is the overall optimization goal, C1, C2, ... are constraints;

[0049] The real-time feedback control mechanism is used to apply the optimization results to the actual operation of the mine tunnel. When the monitoring data changes, the optimization decision is updated in real time based on the multi-objective optimization results, and the support structure parameters X are automatically adjusted according to the optimization results. support , or adjust the mining progress P of the area mining , the expression is:

[0050] ΔX=f(ΔD monitoring ,X previous , …)

[0051] ΔX is the optimized support structure adjustment amount, ΔD monitoring is the change in monitoring data, X previous It is the state of the support structure at the time of the last optimization.

[0052] A deformation monitoring system for mine tunnel operation period, characterized by comprising a real-time data acquisition and transmission module, a deformation prediction and regression analysis module, an anomaly detection and risk warning module, and a dynamic adjustment and optimization control module:

[0053] The real-time data acquisition and transmission module uses sensor equipment to monitor the deformation in the mine tunnel in real time, and transmits the collected monitoring data to the data processing center through the wireless transmission module;

[0054] The deformation prediction and regression analysis module uses an adaptive weighted regression model based on real-time historical data and on-site sensor data. It dynamically adjusts the weight factor according to the historical data of different parts of the tunnel, inputs the collected data into the regression model, and uses a linear fitting method to predict the deformation trend of the tunnel in the future, and updates the prediction results in real time.

[0055] The anomaly detection and risk warning module combines deformation prediction results with an anomaly detection algorithm based on deep learning to analyze monitoring data and identify potential sudden deformation risks. When the prediction results exceed the set safety threshold, the system triggers an early warning signal to remind operators to take emergency measures. At the same time, based on the deformation trend, the risk level and accident probability are evaluated to provide a decision-making basis for the safety management of mine tunnels.

[0056] The dynamic adjustment and optimization control module, with the support of deformation monitoring and early warning mechanism, combines multi-objective optimization algorithm to dynamically adjust the operation of mine tunnels, automatically adjust the parameters in the operation process according to the monitoring results to achieve the optimal safety control strategy, and optimize the operation management of mine tunnels in real time under feedback control.

[0057] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0058] The present invention can greatly improve the real-time and accuracy of mine tunnel deformation monitoring by adopting a real-time monitoring system based on a variety of sensor devices and combining it with advanced data transmission technology. In traditional monitoring methods, sensor data often has problems such as collection lag and transmission delay, resulting in a time difference between the actual deformation situation and the monitoring results, and missing the critical early warning opportunity. However, this system uses a wireless transmission module to quickly transmit the data collected in real time by various sensors to the data processing center, thereby ensuring the timeliness of the data. The data processing center uses an efficient analysis algorithm that can quickly identify deformation trends and provide real-time feedback, greatly shortening the delay from monitoring data to feedback results. In this way, the operators of the mine tunnel can promptly grasp the actual deformation of the tunnel, respond to potential dangers in real time, avoid safety accidents caused by monitoring lags, and improve the reliability of safe operation of the tunnel.

[0059] Through the application of multi-objective optimization algorithm (MOGA), the present invention can comprehensively consider multiple objectives such as the safety, efficiency, and economy of the tunnel with the support of real-time monitoring data, and perform automated safety management and operation optimization. The system does not rely solely on a single safety warning, but develops safety control strategies for different situations through data fusion and algorithm optimization. For example, when a tunnel undergoes local deformation, the system will automatically adjust key parameters such as the strength of the support structure, the mining sequence, or the construction area according to real-time data, geological conditions, and mining progress to achieve the optimal safety control solution. This optimization method based on big data and intelligent algorithms not only improves the scientificity and accuracy of operation management, but also significantly reduces the misjudgment and delays that may be caused by manual intervention. The adaptive adjustment function of the system can also continuously optimize the solution as the operation stage of the mine tunnel changes, ensuring that the best safety control and resource utilization can be achieved at any stage. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] 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.

[0061] Figure 1 The present invention is a flow chart of a method for monitoring deformation during operation of a mine tunnel.

[0062] Figure 2 The present invention is a module schematic diagram of a deformation monitoring system applied to a mine tunnel operation period. DETAILED DESCRIPTION

[0063] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0064] The present invention provides Figure 1 A deformation monitoring method for a mine tunnel during operation is shown, comprising the following steps:

[0065] Use sensor equipment to monitor the deformation in the mine tunnel in real time, including displacement sensors, strain gauges, pressure sensors, etc., and transmit the collected monitoring data to the data processing center through wireless transmission modules;

[0066] The specific steps of using sensor equipment to monitor the deformation in the mine tunnel in real time and transmitting the collected monitoring data to the data processing center through the wireless transmission module are as follows;

[0067] In a mine tunnel, sensors must first be properly deployed based on geological conditions, tunnel structure, and deformation monitoring requirements;

[0068] The selection of sensors is crucial. Common sensors include displacement sensors, strain gauges, pressure sensors, accelerometers, etc. Displacement sensors are used to monitor the lateral and longitudinal deformation of the tunnel, strain gauges are used to measure the strain changes of the tunnel structure, and pressure sensors can be used to monitor the pressure changes of the rock mass. These sensors should be selected according to the geological characteristics of the tunnel and the types of deformation that may occur. Sensors need to be distributed in key locations of the tunnel, such as the vault, side walls, foundations and other areas prone to deformation. The layout of the sensors should cover the entire tunnel as much as possible, and be flexibly arranged according to the deformation characteristics of different areas to ensure that possible local deformations or sudden events can be accurately captured.

[0069] When installing the sensor, ensure that the sensor can work stably and accurately in the tunnel environment;

[0070] The installation method of the sensor needs to be determined according to the structural characteristics of the tunnel and the type of equipment. Displacement sensors usually need to be installed on the key load-bearing structures of the tunnel, while strain gauges are generally installed in areas that are more sensitive to structural deformation, such as the connection points of the support frame or the supporting structure. During the installation process, it is necessary to ensure that the sensor is in good contact with the tunnel surface, and the sensor cable needs to be properly arranged and protected to avoid equipment damage caused by tunnel construction, operation or other external factors. After installation, debugging and verification must be carried out to ensure that the sensor's data acquisition and signal transmission are working properly. The debugging process also needs to confirm the accuracy and response time of the sensor to ensure its reliability in actual operation.

[0071] Once the sensor is installed and debugged successfully, the system will start collecting data in real time;

[0072] Each sensor will continuously monitor physical quantities such as deformation, strain or pressure in the tunnel according to the set sampling frequency. The collected data will be transmitted to the data processing system as soon as possible. During real-time monitoring, the data acquisition system must have a certain intelligent algorithm to automatically identify abnormal data points or sudden deformation events and respond. For example, according to the set threshold, when the sensor detects an abnormality, the system can trigger an alarm in time to notify the staff to conduct inspections or emergency disposal. At this time, the data transmission module (such as the wireless transmission module) will transmit the data measured by the sensor to the data processing center through a wireless network or other transmission methods for further processing and analysis. The stability and reliability of data transmission are crucial, so it is necessary to use high-quality, anti-interference transmission modules and regularly inspect and maintain the communication network.

[0073] After data collection, the sensor sends the data to the data processing center through the wireless transmission module;

[0074] In order to ensure the real-time and reliability of data transmission, the transmission module needs to have efficient data transmission capabilities and be able to operate stably in complex mine tunnel environments. These data are usually sent to the central server and protected by encryption and verification to ensure the security and accuracy of the data during transmission. At the same time, there should be a redundant mechanism in the data transmission process. For example, when a problem occurs in a data transmission path, it can automatically switch to an alternative path to ensure that the data is not lost. After receiving the data, the data processing center will store it in the database for long-term monitoring and subsequent analysis. During the storage process, the data must not only save real-time monitoring data, but also archive historical data for trend analysis, anomaly detection and prediction model training. Data management in this process is very critical, and the integrity and traceability of the data must be ensured to enable long-term monitoring and evaluation of tunnel deformation.

[0075] Based on the historical data and field sensor data collected in real time, an adaptive weighted regression model is used to dynamically adjust the weight factor according to the historical data of different parts of the tunnel. The collected data is input into the regression model, and the deformation trend of the tunnel in the future is predicted using the linear fitting method, and the prediction results are updated in real time.

[0076] Based on the real-time collected historical data and field sensor data, the linear fitting method is used to predict the deformation trend of the tunnel in the future. The specific steps for real-time updating of the prediction results are as follows:

[0077] First, the historical deformation data during the operation period of the mine tunnel is analyzed to provide a data basis for the regression model. The historical data includes the time series of physical quantities such as displacement, strain, and pressure at each monitoring point. These data can be obtained through the long-term monitoring system. The calculation expression is as follows:

[0078] , where D t is the deformation at time point t, X it is the measurement data (such as strain, pressure, etc.) of sensor i at time point t in the historical data, w i is the weight factor of sensor i at time point t, ∈ t is the error introduced in the model prediction, which represents the difference between the actual deformation and the model prediction, and n is the total number of sensors;

[0080] Weighting factor w i In the initial stage, the weight w is set according to the statistical characteristics of historical data (such as deformation rate, regional importance, etc.). i It can be initialized by a simple mean or empirical value, but will be adjusted dynamically based on sensor data later. By building a preliminary regression model, it lays the foundation for subsequent dynamic adjustment.

[0081] With the continuous input of real-time data, it is necessary to adjust the weight factors of each area according to the current monitoring data so that the model can respond more accurately to local sudden deformation. The adaptive weighted regression model is used to dynamically adjust the weight according to the monitored local deformation. The calculation expression is as follows:

[0082] w i (t) = w i (t-1)+η·δ i (t)

[0083] , where w i (t) represents the weight of sensor i at time point t, w i (t-1) is the weight of sensor i at time point t-1, η is the learning rate, which controls the amplitude of weight update, and δ i (t) is the weight change that sensor i needs to adjust at time point t, δ i The calculation formula of (t) is as follows:

[0084] , where ΔD i (t) is the change in deformation of sensor i at time point t, ΔD j (t) is the change in deformation of sensor j at time point t, It is the sum of the deformation changes measured by n sensors at time point t, reflecting the overall deformation trend of the mine tunnel at the current moment;

[0086] By summarizing the deformation changes of all sensors, we can get a rough idea of ​​the deformation of the entire tunnel, which helps to weigh the relative importance of each sensor data. i (t) will increase in the total, and the weight of the sensor w i will be raised so that the model pays more attention to the data of this sensor. This weighting mechanism ensures that the monitoring data of important deformation areas play an important role in the overall prediction.

[0087] The learning rate is a control parameter that determines the step size of the model each time the weight is updated. In the formula, the learning rate (usually denoted as η) is used to control the sensor weight factor w i The update speed of the learning rate. When the learning rate is high, the weights will adjust quickly and respond quickly to changes in monitoring data. However, if it is too large, it may cause drastic changes in the weights and affect stability. On the contrary, if the learning rate is too low, the weights will adjust too slowly, affecting the model's ability to respond to sudden deformations. Therefore, the role of the learning rate here is to find a balance so that the model can adjust the weights of the sensors stably and in a timely manner, and respond sensitively and effectively to sudden changes in the deformation of the mine tunnel.

[0088] After dynamically adjusting the weight factor, the sensor data collected in real time is input into the regression model for deformation prediction. The calculation expression is as follows:

[0089] , where is the predicted value of tunnel deformation at time point t;

[0091] At this stage, the regression model uses a linear fitting method to solve the optimal combination of weight factors through the least squares method to obtain a prediction of future deformation trends. By updating the monitoring data at each time point in real time, the system can timely obtain the changing trend of tunnel deformation and provide a basis for tunnel safety warning.

[0092] With the continuous increase of deformation monitoring data in the tunnel and the continuous optimization of the model, the prediction results also need to be updated in real time to further improve the prediction accuracy of the model. The sliding window method is used to roll the historical data. The calculation expression is as follows:

[0093] , where is the predicted deformation value after the time window T, w i(t-T+1) is the weight factor adjusted within the rolling window, X it-T+1 is the input data within the time window T;

[0095] Combined with the deformation prediction results, the deep learning-based anomaly detection algorithm is used to analyze the monitoring data and identify potential sudden deformation risks. When the prediction results exceed the set safety threshold, the system triggers an early warning signal to remind operators to take emergency measures. At the same time, based on the deformation trend, the risk level and accident probability are evaluated to provide a decision-making basis for the safety management of mine tunnels.

[0096] Combined with the deformation prediction results, the steps to identify potential sudden deformation risks by applying the deep learning-based anomaly detection algorithm are as follows:

[0097] First, the real-time monitoring data must be preprocessed and feature extracted;

[0098] Monitoring data may contain noise or outliers, such as sensor error readings, network transmission errors, etc., which will affect the accuracy of the anomaly detection algorithm. The steps of data preprocessing include denoising, smoothing and normalization. For example, wavelet denoising technology can be used to reduce high-frequency noise, or moving average method can be used to smooth data trends. The preprocessed data will be more stable and continuous, and irrelevant information will be eliminated, which is convenient for subsequent feature extraction. Feature extraction is to extract key variables related to tunnel deformation from the preprocessed data, such as deformation rate, cumulative deformation, acceleration, etc. These features will be used as input to the algorithm to help the deep learning model identify abnormal deformation more accurately.

[0099] Based on the preprocessed data, the anomaly detection model is trained using a deep learning algorithm;

[0100] Commonly used deep learning methods include long short-term memory networks (LSTM), convolutional neural networks (CNN), etc. These models perform well in processing time series data. First, based on normal historical deformation data, an anomaly detection model is trained to let the model learn the pattern of normal deformation trends. The training process can adopt unsupervised or semi-supervised learning methods. By allowing the model to identify the characteristic distribution of normal data, when new data is input, if the model detects a significant difference from the normal pattern, it will be judged as abnormal. Through repeated training and verification, the model can gradually learn the normal change law of tunnel deformation and has high accuracy and robustness. After training, the model will be applied to the detection of real-time monitoring data to identify potential abnormal deformation signals.

[0101] After the deep learning model is deployed in the deformation monitoring system, the real-time monitoring data at each moment is automatically analyzed;

[0102] The deep learning model predicts and judges the data of each sensor to identify whether there is an abnormal change trend. At the same time, in order to avoid false positives or false negatives, a safety threshold is set based on the range of normal deformation data or the safety standard defined by the manager. When the model prediction result exceeds the threshold, it will immediately determine that the deformation data is abnormal and trigger an early warning signal. In addition, information such as the time when the threshold is exceeded, the magnitude of the excess, and the deformation location will be recorded to provide support for subsequent risk assessment. Real-time monitoring and threshold judgment are the core of the entire anomaly detection system. It can quickly identify sudden deformation situations and help the system respond to potential risks in the first place.

[0103] When the monitoring data is judged as abnormal by the model and exceeds the set safety threshold, further risk assessment is carried out;

[0104] Risk assessment consists of two parts: one is to assess the severity of deformation, that is, the magnitude of deformation, acceleration, and the scope of the affected tunnel; the other is to calculate the probability of an accident. Based on historical data, current deformation trends, and the prediction results of the deep learning model, a comprehensive analysis of the possible impact of the current deformation is conducted, and the risks are divided into different levels, such as low risk, medium risk, and high risk. According to the risk level, different early warning response strategies are formulated. For example, in high-risk situations, a red alarm is directly triggered and an emergency notice is sent to the operator, requiring immediate emergency measures; while for medium risks, it may only remind the monitoring personnel to pay close attention. In addition, the risk assessment data will be recorded in the database to provide data support for the long-term safety management of the tunnel. This process realizes the automatic detection, grading, and timely early warning of deformation risks, greatly improving the efficiency and accuracy of mine tunnel safety management.

[0105] With the support of deformation monitoring and early warning mechanism, combined with multi-objective optimization algorithm, the operation of mine tunnels is dynamically adjusted, and the parameters in the operation process are automatically adjusted according to the monitoring results to achieve the optimal safety control strategy, and the operation management of mine tunnels is optimized in real time under feedback control.

[0106] With the support of deformation monitoring and early warning mechanism, combined with multi-objective optimization algorithm, the steps of dynamically adjusting mine tunnel operation are as follows;

[0107] With the support of the deformation monitoring and early warning mechanism of mine tunnels, it is first necessary to integrate and summarize the real-time monitoring data obtained from the sensor network. These data include various physical quantities such as deformation displacement, strain, pressure, etc., covering various key areas in the tunnel. When integrating through the multi-objective optimization algorithm (MOGA), the construction of the objective function is very critical. The objective function is usually composed of multiple sub-objectives, such as safety objectives, economic benefit objectives, and environmental impact objectives. The calculation expression is as follows:

[0108] F=λ1·f safety +λ2·f cost +λ3·f environment

[0109] , where F is the objective function, f safety represents the safety goal, f cost represents the economic benefit target, f environment represents the environmental impact target, λ1, λ2 and λ3 are the safety targets f safety 、Safety goal safety and economic benefit goals cost The weight coefficient of

[0110] In the multi-objective optimization process, in addition to the objective function, corresponding constraints need to be constructed to ensure that the decision-making in the optimization process is reasonable and in line with actual limitations. The tunnel support structure, mining sequence, equipment bearing capacity, etc. may be subject to a series of physical, economic and technical constraints. Assume that the stability force requirement of the support structure is R support , and with the support of monitoring data, ensure R support is greater than the threshold τ, the expression is as follows:

[0111] R support ≥τ

[0112] In addition, the mining sequence should be dynamically adjusted according to the real-time deformation data to ensure the optimal resource scheduling. The expression is as follows:

[0113] P mining ≤P safe (where P safe is a pre-defined safe mining rate)

[0114] , where P mining It indicates the current mining progress of the mine tunnel, usually measured as the mining rate per unit time, or the advancement speed of the mining area, P safe It is a predefined safe mining rate, which is determined based on historical data, geological surveys, engineering design, etc. It represents the maximum mining speed or advancement rate that the tunnel support structure can safely withstand under given geological conditions;

[0115] When executing the multi-objective optimization algorithm, various objectives and constraints are comprehensively analyzed and optimized using methods such as simulated annealing, genetic algorithm or particle swarm optimization. The core of multi-objective optimization is to find a compromise between various objectives by exploring different solution spaces to achieve the global optimal or near-optimal state. Suppose the optimized result of the tunnel support structure is X support , the mining progress is adjusted to P mining, then the optimization problem can be transformed into the following form;

[0116] Maximize F(X support , P mining ,…)subject to C1, C2,…

[0117] , where F(X support , P mining , …) is the overall optimization goal, including indicators such as the stability of the support structure, mining sequence, and equipment efficiency. C1, C2, … are constraints, which cover restrictions in terms of safety, resource utilization, and economic benefits. By dynamically adjusting key parameters, the operation strategy of the tunnel is adaptively optimized according to real-time monitoring data, such as adjusting the thickness of the support structure, reinforcing the area, or adjusting the mining sequence, to ensure that the tunnel is always in the optimal safety state during construction or operation.

[0118] The real-time feedback control mechanism is used to apply the optimization results to the actual operation of the mine tunnel. When the monitoring data changes, the optimization decision is updated in real time based on the multi-objective optimization results. The monitoring data shows that a certain part of the tunnel has a large deformation. The support structure parameters X are automatically adjusted according to the optimization results. support , or adjust the mining progress P of the area mining , the expression is:

[0119] ΔX=f(ΔD monitoring ,X previous , …)

[0120] ΔX is the adjustment amount of the optimized support structure, ΔD monitoring is the change in monitoring data, X previous is the state of the support structure at the last optimization;

[0121] Implementation method one: During the operation of mine tunnels, the reliability of the real-time deformation monitoring and early warning system is crucial. In order to ensure the stability and safety of the tunnel during long-term operation, this implementation method monitors key locations in the tunnel in real time by deploying a variety of sensors (such as displacement sensors, strain gauges, pressure sensors, etc.) to obtain data such as tunnel deformation, strain, and pressure. These sensors are usually deployed in key locations such as the tunnel's support structure, tunnel walls, and bottom to ensure that the overall operating status of the tunnel can be fully reflected. All data collected by the sensors are sent to the data processing center in real time through a wireless transmission module.

[0122] After receiving these data, the data processing center will conduct a comprehensive analysis of the monitoring data through a multi-objective optimization algorithm (MOGA). The system uses this algorithm to evaluate the different safety, economic, and efficiency goals of the tunnel. Each goal has a different priority, such as tunnel safety taking precedence over economic benefits, or maximizing resource mining efficiency while ensuring safety. The system generates the optimal operation adjustment plan based on the real-time collected data, combined with the geological conditions, environmental factors, and mining conditions of the tunnel. For example, when the displacement detected at a certain monitoring point exceeds the set safety threshold, the system will immediately trigger an early warning and recommend strengthening the support structure or adjusting the mining sequence to reduce the risk of deformation.

[0123] The advantage of this system is that it is highly automated and intelligent. It can not only perceive the state changes of the tunnel in real time, but also flexibly adjust the monitoring parameters and emergency measures according to different safety requirements. By dynamically analyzing the deformation of the tunnel, the system can actively adjust key parameters such as support strength and mining progress to ensure the safe operation of the tunnel in a complex geological environment. At the same time, the efficiency and accuracy of data transmission and processing also greatly reduce the errors caused by manual intervention and avoid potential accidents caused by monitoring lag or data loss.

[0124] In addition, the system's optimization algorithm can adapt to different stages of mine tunnel operation and dynamically adjust monitoring strategies and emergency response measures. For example, in the early stages of tunnel construction, more attention may be paid to geological pressure and the stability of the support structure, while in the mining stage, the monitoring focus may shift to changes such as displacement and settlement. As the system continues to accumulate and analyze historical data, its optimization model can gradually improve its prediction capabilities and processing accuracy, so that it can provide accurate and real-time decision support when facing complex and sudden mining environments.

[0125] Implementation method 2: This implementation method focuses on designing a dynamic feedback control and adaptive adjustment mechanism to ensure the safety and efficiency of the tunnel in operation. During the operation of a mine tunnel, factors such as changes in geological conditions, rock mass movement, and construction progress often cause the deformation of the tunnel to change continuously. Therefore, how to quickly respond to changes in deformation data and make timely and effective adjustments to the tunnel's operating strategy is the core function of this system.

[0126] First, the system uses various sensors to monitor the displacement, strain, pressure and other key data in the tunnel in real time, and the data is wirelessly transmitted to the data processing center. The data processing center processes and analyzes these real-time data through advanced analysis algorithms, and combines the historical operation data of the tunnel to evaluate the current safety status of the tunnel. If the deformation speed or stress value of a certain part exceeds the set safety standard, the system will immediately issue an alarm and calculate the best response measures. These measures may include strengthening the strength of the support structure, adjusting the mining sequence or changing the construction area, and the specific adjustment plan will be automatically generated according to the current tunnel status.

[0127] A key advantage of the dynamic feedback control mechanism is its high adaptability. Whenever the monitoring data changes, the system will immediately recalculate and adjust the optimization strategy. This process is real-time, and the system dynamically adjusts key parameters in tunnel operation, such as support strength and mining progress, according to actual conditions to ensure that it can respond to possible risks at any time. In practical applications, this mechanism not only improves the speed of emergency response, but also reduces errors in manual intervention, especially in the case of sudden changes, and can quickly respond to the latest data.

[0128] In addition to responding to real-time data, the system also has the ability to self-learn. As the mine tunnel operation data accumulates, the system will continue to learn and optimize the algorithm, and improve the prediction accuracy and decision-making efficiency through the review and analysis of historical data. For example, when the system encounters a certain type of special geological conditions, it will automatically adjust the parameter settings and optimize the subsequent monitoring and control strategies to ensure that the safety and stability of tunnel operations can be maintained even under complex geological conditions.

[0129] The core advantage of the system lies in its efficient real-time and adaptive capabilities. It can flexibly adjust the tunnel's operating strategy according to the ever-changing operating data and environmental factors, ensuring that the tunnel can always maintain the optimal safety state in a complex and dynamic environment. This system significantly improves the risk control level of mine tunnels during operation and reduces safety hazards caused by geological changes or improper mining progress.

[0130] Implementation method three: This implementation method designs a comprehensive early warning system and emergency response linkage mechanism to further improve the safety and emergency handling capabilities of mine tunnels. During operation, mine tunnels often face high safety risks due to factors such as external environment and internal deformation. In order to effectively avoid accidents, the system can detect problems and take measures in time before potential risks occur through real-time deformation monitoring, intelligent data analysis and optimization calculation, ensuring that mine tunnels can take appropriate preventive measures in the shortest time.

[0131] The system first relies on various sensors (such as displacement sensors, strain gauges, pressure sensors, etc.) deployed in the tunnel to monitor the status of the tunnel in real time. The data of these sensors is quickly transmitted to the data processing center through the wireless transmission module. The system analyzes the real-time data and generates a safety status report. If the system detects that the deformation speed of a certain area suddenly increases or reaches the safety threshold, the early warning system will immediately issue an alarm and recommend emergency response measures through an automated decision-making model. These emergency measures may include: suspending mining, immediately reinforcing the support structure, or adjusting the tunnel mining sequence, etc., to reduce the expansion of potential risks.

[0132] The system can provide tailor-made emergency plans for different types of risks through an integrated multi-objective optimization algorithm. For example, if the system finds that the strain value in a certain area has increased sharply, it will calculate the most effective emergency response plan based on the geological characteristics, tunnel structure and deformation trend of the area, and dispatch equipment and personnel for emergency treatment in advance. The early warning system can not only identify potential safety hazards, but also provide targeted emergency plans in a timely manner, helping managers make correct decisions at the first time, thereby minimizing accidents.

[0133] In addition, the emergency response mechanism also combines the actual situation of mine tunnels to provide workers with clear operating guidelines and emergency measures in emergency situations. For example, if local displacement abnormalities occur during tunnel mining, the system will automatically calculate and recommend emergency treatment steps based on the optimization model, including whether to suspend mining, whether to strengthen support, whether to change the working face, etc. The emergency response mechanism can not only improve processing efficiency, but also minimize potential losses caused by human judgment errors.

[0134] The core advantage of this linkage mechanism is that it can detect risks in advance through an intelligent early warning system and provide the best response plan based on actual conditions. By integrating monitoring, early warning and emergency response, the system ensures the safe operation of mine tunnels, significantly improves risk management capabilities and emergency response efficiency, and thus provides more reliable safety guarantees for mining companies.

[0135] The present invention can greatly improve the real-time and accuracy of mine tunnel deformation monitoring by adopting a real-time monitoring system based on a variety of sensor devices and combining it with advanced data transmission technology. In traditional monitoring methods, sensor data often has problems such as collection lag and transmission delay, resulting in a time difference between the actual deformation situation and the monitoring results, and missing the critical early warning opportunity. However, this system uses a wireless transmission module to quickly transmit the data collected in real time by various sensors to the data processing center, thereby ensuring the timeliness of the data. The data processing center uses an efficient analysis algorithm that can quickly identify deformation trends and provide real-time feedback, greatly shortening the delay from monitoring data to feedback results. In this way, the operators of the mine tunnel can promptly grasp the actual deformation of the tunnel, respond to potential dangers in real time, avoid safety accidents caused by monitoring lags, and improve the reliability of safe operation of the tunnel.

[0136] Through the application of multi-objective optimization algorithm (MOGA), the present invention can comprehensively consider multiple objectives such as the safety, efficiency, and economy of the tunnel with the support of real-time monitoring data, and perform automated safety management and operation optimization. The system does not rely solely on a single safety warning, but develops safety control strategies for different situations through data fusion and algorithm optimization. For example, when a tunnel undergoes local deformation, the system will automatically adjust key parameters such as the strength of the support structure, the mining sequence, or the construction area according to real-time data, geological conditions, and mining progress to achieve the optimal safety control solution. This optimization method based on big data and intelligent algorithms not only improves the scientificity and accuracy of operation management, but also significantly reduces the misjudgment and delays that may be caused by manual intervention. The adaptive adjustment function of the system can also continuously optimize the solution as the operation stage of the mine tunnel changes, ensuring that the best safety control and resource utilization can be achieved at any stage.

[0137] The present invention provides Figure 2 The deformation monitoring system shown is applied to the operation period of a mine tunnel, including a real-time data acquisition and transmission module, a deformation prediction and regression analysis module, an anomaly detection and risk warning module, and a dynamic adjustment and optimization control module:

[0138] The real-time data acquisition and transmission module uses sensor equipment to monitor the deformation in the mine tunnel in real time, and transmits the collected monitoring data to the data processing center through the wireless transmission module;

[0139] The deformation prediction and regression analysis module uses an adaptive weighted regression model based on real-time historical data and on-site sensor data. It dynamically adjusts the weight factor according to the historical data of different parts of the tunnel, inputs the collected data into the regression model, and uses a linear fitting method to predict the deformation trend of the tunnel in the future, and updates the prediction results in real time.

[0140] The anomaly detection and risk warning module combines deformation prediction results with an anomaly detection algorithm based on deep learning to analyze monitoring data and identify potential sudden deformation risks. When the prediction results exceed the set safety threshold, the system triggers an early warning signal to remind operators to take emergency measures. At the same time, based on the deformation trend, the risk level and accident probability are evaluated to provide a decision-making basis for the safety management of mine tunnels.

[0141] The dynamic adjustment and optimization control module, with the support of deformation monitoring and early warning mechanism, combines multi-objective optimization algorithm to dynamically adjust the operation of mine tunnels, automatically adjust the parameters in the operation process according to the monitoring results to achieve the optimal safety control strategy, and optimize the operation management of mine tunnels in real time under feedback control.

[0142] An embodiment of the present invention provides a method for monitoring deformation during the operation period of a mine tunnel, which is realized by the above-mentioned system for monitoring deformation during the operation period of a mine tunnel. The specific method and process of the system for monitoring deformation during the operation period of a mine tunnel are detailed in the above-mentioned embodiment of a method for monitoring deformation during the operation period of a mine tunnel, which will not be repeated here.

[0143] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A deformation monitoring method for mine tunnel operation period, characterized in that: The following steps are involved: Use sensor equipment to monitor the deformation in the mine tunnel in real time, and transmit the collected monitoring data to the data processing center through the wireless transmission module; Based on the historical data and field sensor data collected in real time, an adaptive weighted regression model is used to dynamically adjust the weight factor according to the historical data of different parts of the tunnel. The collected data is input into the regression model, and the deformation trend of the tunnel in the future is predicted using the linear fitting method, and the prediction results are updated in real time. Combined with the deformation prediction results, the deep learning-based anomaly detection algorithm is used to analyze the monitoring data and identify potential sudden deformation risks. When the prediction results exceed the set safety threshold, the system triggers an early warning signal to remind operators to take emergency measures. At the same time, based on the deformation trend, the risk level and accident probability are evaluated to provide a decision-making basis for the safety management of mine tunnels. With the support of deformation monitoring and early warning mechanism, combined with multi-objective optimization algorithm, the operation of mine tunnels is dynamically adjusted, and the parameters in the operation process are automatically adjusted according to the monitoring results to achieve the optimal safety control strategy, and the operation management of mine tunnels is optimized in real time under feedback control.

2. The method for monitoring deformation during operation of a mine tunnel according to claim 1 is characterized in that: The specific steps of using sensor equipment to monitor the deformation in the mine tunnel in real time and transmitting the collected monitoring data to the data processing center through the wireless transmission module are as follows; In a mine tunnel, sensors must first be properly deployed based on geological conditions, tunnel structure, and deformation monitoring requirements; When installing the sensor, ensure that the sensor can work stably and accurately in the tunnel environment; Once the sensor is installed and debugged successfully, the system will start collecting data in real time; After data collection, the sensor sends the data to the data processing center through the wireless transmission module.

3. The method for monitoring deformation during operation of a mine tunnel according to claim 1, characterized in that: Based on the real-time collected historical data and field sensor data, the linear fitting method is used to predict the deformation trend of the tunnel in the future. The specific steps for real-time updating of the prediction results are as follows: First, the historical deformation data of the mine tunnel during operation is analyzed to provide a data basis for the regression model. The calculation expression is as follows: , where D t is the deformation at time point t, X it is the measurement data of sensor i at time point t in the historical data, w i is the weight factor of sensor i at time point t, ε t is the error introduced in the model prediction, which represents the difference between the actual deformation and the model prediction, and n is the total number of sensors; With the continuous input of real-time data, it is necessary to adjust the weight factors of each area according to the current monitoring data so that the model can respond more accurately to local sudden deformation. The adaptive weighted regression model is used to dynamically adjust the weight according to the monitored local deformation. The calculation expression is as follows: w i (t) = w i (t-1)+η·δ i (t), where w i (t) represents the weight of sensor i at time point t, w i (t-1) is the weight of sensor i at time point t-1, η is the learning rate, which controls the amplitude of weight update, and δ i (t) is the weight change that sensor i needs to adjust at time point t, δ i The calculation formula of (t) is as follows: , where ΔD i (t) is the change in deformation of sensor i at time point t, ΔD j (t) is the change in deformation of sensor j at time point t, It is the sum of the deformation changes measured by n sensors at time point t, reflecting the overall deformation trend of the mine tunnel at the current moment.

4. The method for monitoring deformation during operation of a mine tunnel according to claim 3 is characterized in that: After dynamically adjusting the weight factor, the sensor data collected in real time is input into the regression model for deformation prediction. The calculation expression is as follows: , where is the predicted value of tunnel deformation at time point t; With the continuous increase of deformation monitoring data in the tunnel and the continuous optimization of the model, the prediction results also need to be updated in real time to further improve the prediction accuracy of the model. The sliding window method is used to roll the historical data. The calculation expression is as follows: , where is the predicted deformation value after the time window T, w i (t-T+1) is the weight factor adjusted within the rolling window, X it-T+1 is the input data within the time window T.

5. The method for monitoring deformation during operation of a mine tunnel according to claim 1, characterized in that: Combined with the deformation prediction results, the steps to identify potential sudden deformation risks by applying the deep learning-based anomaly detection algorithm are as follows: First, the real-time monitoring data must be preprocessed and feature extracted; Based on the preprocessed data, the anomaly detection model is trained using a deep learning algorithm; After the deep learning model is deployed in the deformation monitoring system, the real-time monitoring data at each moment is automatically analyzed; When the monitoring data is judged as abnormal by the model and exceeds the set safety threshold, further risk assessment is performed.

6. The method for monitoring deformation during operation of a mine tunnel according to claim 1 is characterized in that: With the support of deformation monitoring and early warning mechanism, combined with multi-objective optimization algorithm, the steps of dynamically adjusting mine tunnel operation are as follows; With the support of the deformation monitoring and early warning mechanism of mine tunnels, it is first necessary to integrate and summarize the real-time monitoring data obtained from the sensor network. The calculation expression is as follows: F=λ1·f safety +λ2·f cost +λ3·f environment , where F is the objective function, f safety represents the safety goal, f cost represents the economic benefit target, f environment represents the environmental impact target, λ1, λ2 and λ3 are the safety targets f safety 、Safety goal safety and economic benefit goals cost The weight coefficient of In the multi-objective optimization process, in addition to the objective function, corresponding constraints need to be constructed to ensure that the decision in the optimization process is reasonable and meets the actual constraints. Assume that the stability force requirement of the support structure is R support , and with the support of monitoring data, ensure R support is greater than the threshold τ, the expression is as follows: R support ≥τ In addition, the mining sequence should be dynamically adjusted according to the real-time deformation data to ensure the optimal resource scheduling. The expression is as follows: P mining ≤P safe (where P safe is a pre-defined safe mining rate), where, P mining Indicates the current mining progress of the mine tunnel, P safe is a predefined safe mining rate; When executing the multi-objective optimization algorithm, various objectives and constraints are comprehensively analyzed and optimized using methods such as simulated annealing, genetic algorithm or particle swarm optimization. Suppose the optimization result of the tunnel support structure is X support , the mining progress is adjusted to P mining , then the optimization problem can be transformed into the following form; MaximizeF(X support , P mining , ...) subject to C1, C2, ..., where F(X support , P mining , ...) is the overall optimization goal, C1, C2, ... are constraints; The real-time feedback control mechanism is used to apply the optimization results to the actual operation of the mine tunnel. When the monitoring data changes, the optimization decision is updated in real time based on the multi-objective optimization results, and the support structure parameters X are automatically adjusted according to the optimization results. support , or adjust the mining progress P of the area mining , the expression is: ΔX=f(ΔD monitoring ,X previous ,……) ΔX is the optimized support structure adjustment amount, ΔD monitoring is the change in monitoring data, and Xprevious is the state of the support structure at the last optimization.

7. A deformation monitoring system for a mine tunnel during operation, used to implement a deformation monitoring method for a mine tunnel during operation as described in any one of claims 1 to 6, characterized in that: It includes real-time data acquisition and transmission module, deformation prediction and regression analysis module, anomaly detection and risk warning module, and dynamic adjustment and optimization control module: The real-time data acquisition and transmission module uses sensor equipment to monitor the deformation in the mine tunnel in real time, and transmits the collected monitoring data to the data processing center through the wireless transmission module; The deformation prediction and regression analysis module uses an adaptive weighted regression model based on real-time historical data and on-site sensor data. It dynamically adjusts the weight factor according to the historical data of different parts of the tunnel, inputs the collected data into the regression model, and uses the linear fitting method to predict the deformation trend of the tunnel in the future, and updates the prediction results in real time. The anomaly detection and risk warning module combines deformation prediction results with an anomaly detection algorithm based on deep learning to analyze monitoring data and identify potential sudden deformation risks. When the prediction results exceed the set safety threshold, the system triggers an early warning signal to remind operators to take emergency measures. At the same time, based on the deformation trend, the risk level and accident probability are evaluated to provide a decision-making basis for the safety management of mine tunnels. The dynamic adjustment and optimization control module, with the support of deformation monitoring and early warning mechanism, combines multi-objective optimization algorithm to dynamically adjust the operation of mine tunnels, automatically adjust the parameters in the operation process according to the monitoring results to achieve the optimal safety control strategy, and optimize the operation management of mine tunnels in real time under feedback control.

Images