Tunnel construction real-time monitoring system based on big data

Abstract

The invention relates to the technical field of construction monitoring, in particular to a tunnel construction real-time monitoring system based on big data. Comprising a server, a real-time monitoring module and a safety early warning module, comprehensive analysis is carried out by fusing multiple parameters such as stress, temperature, tunnel depth and accumulation of micro-seismic events instead of single-dimensional data, and the accuracy of rockburst risk prediction is improved through a multi-dimensional fusion analysis mode; meanwhile, the observation time window length of multi-dimensional data analysis is dynamically adjusted according to the rockburst risk of each area, and data support is provided for safety in the tunneling process; by performing curve fitting on the rockburst risk index change in the observation time window, the risk change trend of each region can be dynamically monitored, the region with the rising risk can be accurately identified, early warning can be given in advance, and refuge route planning can be performed, so that rapid refuge under the rockburst risk is realized, and the safety of tunnel constructors is guaranteed to the greatest extent.

Description

Technical Field

[0001] The present invention relates to the technical field of construction monitoring, and in particular to a real-time monitoring system for tunnel construction based on big data. Background Art

[0002] Rockburst is one of the common geological disasters in the process of tunnel excavation, usually occurring in hard rock mass in deep tunnels or high-stress environments. Rockburst refers to the phenomenon of sudden rock cracking, breaking and splashing due to the high ground stress of the surrounding rock during tunnel excavation, accompanied by strong sounds and vibrations, sometimes similar to small explosions. The root cause is the sudden release of energy by rocks under high stress; therefore, real-time monitoring during tunnel excavation is very necessary to warn of rockbursts.

[0003] At present, the rockburst early warning for real-time monitoring of tunnel construction mainly consists of two methods: single early warning method and multi-factor early warning method. Among them, the single early warning method mainly makes static judgments based on rock characteristics such as rock mass stress intensity, and the quantitative threshold is far from the actual rockburst location; while the multi-factor early warning method mainly focuses on a certain stage from the beginning of rockburst development to the realization of macroscopic damage, and cannot achieve comprehensive early warning of the entire process from scratch, resulting in deviations in early warning accuracy. Therefore, in order to improve the accuracy of rockburst early warning, a more comprehensive and dynamic monitoring and early warning method is needed to ensure the safety of tunnel construction. Summary of the invention

[0004] The purpose of the present invention is to provide a real-time monitoring system for tunnel construction based on big data to solve the problems mentioned in the above background technology.

[0005] The object of the present invention can be achieved by the following technical solutions: A real-time monitoring system for tunnel construction based on big data, comprising: a server, a real-time monitoring module and a safety early warning module;

[0006] The entire tunnel is divided into several areas, each area has several monitoring points, each monitoring point is equipped with various sensors, and the server communicates with various sensors in each area to collect tunnel information of each area, where the tunnel information includes stress, temperature and tunnel depth;

[0007] The real-time monitoring module obtains the risk index within the observation time window based on the fusion analysis of multi-dimensional factors; the risk index is divided by the set time window constant to obtain the latest observation time window, thereby obtaining the latest observation time window of each region, and each region performs multi-dimensional factor fusion analysis according to the latest observation time window;

[0008] The latest observation time window of the safety warning module is based on the risk index of each area in each observation time window to perform development trend analysis to determine whether to generate a rock burst warning. If a rock burst warning is triggered, a safety warning will be issued to the construction personnel inside the tunnel; specifically:

[0009] The received observation time windows are numbered according to their corresponding time sequence, with the sequence number of the observation time window as the horizontal coordinate and the risk index as the vertical coordinate. The risk index is input into the coordinate axis according to its corresponding number to obtain the index point, and a smooth curve is used to connect the index points in sequence to obtain a risk index change curve chart; according to the risk index change curve chart, a development trend analysis is performed to obtain a rock burst warning value;

[0010] The rockburst warning value is compared and analyzed with the set warning threshold. When the rockburst warning value of any area is greater than the set warning threshold, a safety warning is issued to the tunnel construction personnel, and the area corresponding to the rockburst warning value is recorded as a risk area; thus, one or more risk areas can be obtained.

[0011] When a security warning is triggered, the specific warning steps are as follows:

[0012] By communicating with the smart devices of each construction worker in the tunnel, the location of each construction worker can be obtained in real time and marked on the tunnel map;

[0013] It is assumed that there are several refuge points inside the tunnel, and they are marked on the tunnel map; the location of the construction workers inside the tunnel is taken as the starting point, and the nearest refuge point is selected as the end point. The path planning algorithm is used to obtain several paths leading to the nearest refuge points, and each path is marked on the tunnel map to calculate the areas involved in each path and the length of the road section involved; any one of the paths is selected, and the area involved in the path is recorded as the relevant area and the corresponding section length; the rock burst warning value of the relevant area is retrieved, and it is normalized with the section length and its number is taken, and the value is weighted to calculate the section risk value; the section risk value of each relevant area of ​​the path can be obtained, and the sum of the values ​​is calculated to obtain the path risk value, thereby obtaining the path risk values ​​of several paths, and the path with the smallest path risk value is selected as the target path of the construction worker, and it is sent to the construction worker's smart device.

[0014] Preferably, the specific steps of performing development trend analysis based on the risk index change curve are:

[0015] Draw a tangent line at each index point, and use the least squares method to calculate the slope of the tangent line to obtain the trend slope of each index point, which is recorded as Pa; where a = 1, 2, 3...A, A is a positive integer, A is the total number of observation time windows, and a is the serial number of any observation time window;

[0016] The trend slopes greater than zero are summed up to get the exponential increase value, recorded as H1, and the trend slopes less than zero are summed up and the absolute value is taken to get the exponential decrease value, recorded as H2;

[0017] Substitute the risk index FCa, trend slope Pa, index increase value H1 and index decrease value H2 of each observation time window into the set formula The rockburst warning value PH is calculated, where β3 and β4 are the set weight constants. is the mean trend slope of each index point.

[0018] Preferably, the specific steps based on multi-dimensional factor fusion analysis are:

[0019] 301: Assume that there is an observation time window; divide the tunnel information at each acquisition time in the region by the observation time window segment, thereby obtaining the tunnel information in each observation time window; wherein the tunnel information includes the stress, temperature and tunnel depth of each monitoring point;

[0020] 302: Perform stress concentration analysis based on the stress of each monitoring point in the region to obtain a stress concentration value, thereby obtaining the stress concentration value at each acquisition time;

[0021] 303: Based on the temperature of each monitoring point in the area, a stress concentration analysis is performed to obtain the temperature concentration value, thereby obtaining the temperature concentration value at each acquisition time:

[0022] 304: Perform tunnel depth analysis based on the tunnel depths of each monitoring point in the area to determine the overall depth of each area, thereby obtaining the overall depth at each acquisition time;

[0023] 305: Performing microseismic event accumulation analysis based on the number of microseismic events to obtain a microseismic event accumulation constant;

[0024] 306: Calculate the stress concentration value YZi, the temperature concentration value TZi and the overall depth Ri at each acquisition time in the observation time window by the set formula Bi=k1×YZi+k2×TZi+k3×Ri to obtain the rock burst risk value Bi corresponding to each acquisition time, wherein k1, k2, k3 are respectively set weight constants, i=1, 2, 3...I, I is a positive integer, I is the total number of acquisition times in the observation time window, and i is the sequence number of any acquisition time in the observation time window; construct a two-dimensional rectangular coordinate system with time as the abscissa and risk value as the ordinate, input each rock burst risk value in the observation time window into the coordinate system according to the sequence of the corresponding acquisition times, and record the position of the rock burst risk value in the coordinate axis as a risk point, and use a smooth curve to connect each risk point in sequence to obtain a curve graph of rock burst risk value changing with time;

[0025] 307: Draw a tangent line of the curve at each risk point, and use the least squares method to calculate the slope of each tangent line and record it as the risk slope Fi; sum the risk slopes greater than zero to obtain the risk increase value and record it as C1, and sum the risk slopes less than zero and take the absolute value to calculate the risk reduction value and record it as C2; substitute the microseismic event cumulative constant μ, the rock burst risk value Bi at each acquisition time, the risk slope Fi, the risk increase value C1 and the risk reduction value C2 into the set formula The risk index FC of the observation time window is calculated, where β1, β2, and β3 are the set weight constants. is the mean risk slope of each risk point.

[0026] Preferably, the stress concentration analysis:

[0027] The stress of each monitoring point is recorded as Yj, j=1, 2, 3...J, J is a positive integer, J is the total number of monitoring points in the region, and j is the serial number of any monitoring point in the region; the maximum stress and the minimum stress are selected from the stresses corresponding to each monitoring point, and they are recorded as Ymax and Ymin respectively; the stress of each monitoring point is compared and analyzed with the set stress interval, and when the stress is greater than the maximum value in the set stress interval, the monitoring point corresponding to the stress is recorded as a strong stress point, thereby obtaining several strong stress points in the region, and calculating the distance between two adjacent strong stress points respectively as strong stress intervals, thereby obtaining several strong stress intervals, and calculating the average of the strong stress intervals in the region as LY;

[0028] Substitute the stress Yj, maximum stress Ymax, minimum stress Ymin and strong stress interval mean LY of each monitoring point into the set formula The stress concentration value YZj is calculated, where λ1, λ2, and λ3 are set weight constants respectively. is the mean stress value of each monitoring point in the area.

[0029] Preferably, the temperature is concentrated on analyzing:

[0030] The temperature of each monitoring point is recorded as Tj, and the maximum temperature and the minimum temperature are selected from the temperatures corresponding to each monitoring point, and recorded as Tmax and Tmin; the temperature of each monitoring point in the region is compared and analyzed with the set temperature range. When the temperature is greater than the maximum value in the set temperature range, the monitoring point corresponding to the temperature is recorded as a high temperature point; the distance between two adjacent high temperature points is calculated and recorded as a high temperature interval, thereby obtaining several high temperature intervals, and the average of the high temperature intervals is calculated and recorded as LT;

[0031] Substitute the temperature Tj, maximum temperature Tmax, minimum temperature Tmin and high temperature interval mean LT of the monitoring point into the set formula The temperature concentration value is calculated, where λ4, λ5, and λ6 are the set weight constants. is the average temperature of each monitoring point in the area.

[0032] Preferably, tunnel depth analysis:

[0033] Compare and analyze the tunnel depth of each monitoring point with the set depth interval. When the tunnel depth is greater than the maximum value in the set depth interval, the monitoring point corresponding to the tunnel depth is recorded as a depth-exceeding point; when the tunnel depth is within the set depth interval, the monitoring point corresponding to the tunnel depth is recorded as a depth-centering point; when the tunnel depth is less than the minimum value in the set depth interval, the monitoring point corresponding to the tunnel depth is recorded as a depth-upper point; the number of depth-exceeding points, depth-centering points, and depth-upper points in the area is counted respectively, and they are recorded as S1, S2, and S3 respectively;

[0034] When S1≥S2+S3, the tunnel depths corresponding to the depth-exceeding points are averaged as the overall depth of the area; when S3≥S1+S2, the tunnel depths corresponding to the upper depth points are averaged as the overall depth of the area; in other cases, the tunnel depths of each monitoring point in the area are averaged as the overall depth of the area; thus, the overall depth of the area at any acquisition time within the observation time window can be obtained.

[0035] Preferably, cumulative analysis of microseismic events:

[0036] The number of microseismic events and the position, amplitude and energy value corresponding to each microseismic event are obtained; the amplitude and energy are weighted to obtain the microseismic intensity value, and the value is multiplied by the set distance coefficient to obtain the position influence distance; the position of the microseismic event is taken as the center point, and a circle is drawn with the position influence distance as the radius, and the area within the circle is recorded as the influence area, and the center point of each influence area is obtained by using the centroid algorithm, and the distance between the center point and the position of the microseismic event is calculated and recorded as the microseismic distance of the influence area; the microseismic intensity is divided by the microseismic distance to obtain the influence intensity, thereby obtaining the influence intensity of each microseismic event on the influence area, and the sum is calculated to obtain the microseismic event cumulative constant of all microseismic events on the influence area, thereby obtaining the microseismic event cumulative constant of each area.

[0037] Beneficial effects of the present invention:

[0038] 1. By integrating multiple parameters such as stress, temperature, tunnel depth, accumulation of microseismic events, etc. for comprehensive analysis, rather than single-dimensional data, the multi-dimensional fusion analysis method improves the accuracy of rock burst risk prediction and can identify potential high-risk areas; at the same time, the observation time window length of the multi-dimensional data analysis is dynamically adjusted according to the rock burst risk of each area. When the rock burst risk increases, the observation time window will be shortened, so that abnormal changes in the tunnel can be captured more timely and a faster response can be achieved. This dynamic adjustment mechanism improves the flexibility and response speed of tunnel monitoring; it helps engineers take preventive measures in advance during tunnel excavation, reduces the possibility of sudden rock burst accidents, and provides data support for achieving safety in the process of tunnel excavation;

[0039] 2. By fitting a curve to the changes in the rockburst risk index within the observation time window and analyzing it based on the trend slope, the system can dynamically monitor the risk change trend of each area, accurately identify areas with increased risks, and issue warnings in advance to prevent construction workers from entering high-risk areas; when a rockburst warning is triggered, the Dijkstra and other path planning algorithms are used to plan the path from the construction worker's location to the nearest refuge point. By combining the rockburst warning values ​​of different paths, the length of the road section and other factors, a comprehensive assessment of the safety of each path is conducted, and the refuge path with the lowest risk can be selected for each construction worker, significantly improving the efficiency and safety of risk avoidance; so as to achieve rapid refuge under rockburst risk and maximize the safety of tunnel construction workers. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The present invention will be further described below in conjunction with the accompanying drawings.

[0041] Figure 1 It is a schematic diagram of system module connection of the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0043] See also Figure 1 As shown, the present invention is a real-time monitoring system for tunnel construction based on big data, comprising: a server, a real-time monitoring module and a safety early warning module;

[0044] The entire tunnel is divided into several areas (geological conditions, stress distribution and construction progress in different areas of the tunnel may be different, so dividing the tunnel into several areas is helpful for targeted monitoring of the specific conditions of different areas). Each area has several monitoring points, and each monitoring point is equipped with various sensors (such as stress sensors, temperature sensors, etc.). The server communicates with various sensors in each area to collect tunnel information in each area, where the tunnel information includes stress, temperature and tunnel depth;

[0045] The real-time monitoring module conducts real-time monitoring and analysis based on multi-dimensional tunnel information to determine the rockburst risk in each area and generate a risk index for each area accordingly; specifically:

[0046] Step 1: Set an observation time window (the observation time window in the initial state of the system is usually given a value by personnel in this field according to the actual needs of tunnel excavation, such as 30 minutes, 50 minutes, or 60 minutes, etc.); divide the tunnel information at each collection time in the area according to the observation time window, thereby obtaining the tunnel information in each observation time window;

[0047] Step 2: Select any one of the observation time windows, retrieve the tunnel information corresponding to the observation time window, and perform a multi-dimensional fusion analysis of the region from the perspective of stress, temperature, and tunnel depth to determine the rockburst risk status of each area of ​​the tunnel within each observation time window; the tunnel information includes the stress, temperature, and tunnel depth of each monitoring point; specifically:

[0048] Stress concentration analysis: The stress of each monitoring point is recorded as Yj, j = 1, 2, 3...J, J is a positive integer, J is the total number of monitoring points in the region, and j is the serial number of any monitoring point in the region;

[0049] The maximum stress value and the minimum stress value in each monitoring point are selected and recorded as Ymax and Ymin respectively; the stress of each monitoring point is compared and analyzed with the set stress interval. When the stress is greater than the maximum value in the set stress interval, the monitoring point corresponding to the stress is recorded as a strong stress point. Thus, several strong stress points in the region are obtained, and the distance between two adjacent strong stress points is calculated and recorded as the strong stress interval. Thus, several strong stress intervals are obtained, and the average is calculated to obtain the mean of the strong stress interval in the region and recorded as LY. It should be noted that the smaller the mean of the strong stress interval is, the more concentrated the stress is in the region. Using the set formula The stress concentration value YZj is calculated, where λ1, λ2, and λ3 are respectively set weight constants (the weight constants are set by personnel in this field according to the actual situation of tunnel excavation). is the stress mean of each monitoring point in the region; it should be noted that if the stress distribution is more concentrated, the stress concentration value is smaller; the greater the stress of each monitoring point, the greater the stress concentration value; thus, the stress concentration value YZ i corresponding to any acquisition time in the observation window can be obtained; i = 1, 2, 3...I, I is a positive integer, I is the total number of acquisition times in the observation window, and i is the sequence number of any acquisition time in the observation window;

[0050] Temperature concentration analysis: retrieve the temperature of each monitoring point and record it as Tj; it should be noted that the temperature difference between the monitoring points is significant, forming a large temperature gradient, which will lead to the generation of thermal stress and stress concentration, and is one of the signals of increased rock burst risk; increased temperature tends to reduce the strength of rock, especially in high temperature environments, the compressive strength and tensile strength of rock will decrease, resulting in a weakening of its bearing capacity; in this case, the originally stable rock mass will undergo rock burst due to stress exceeding its reduced strength, increasing the risk of rock burst, so temperature monitoring cannot be ignored; select the maximum temperature and minimum temperature at each monitoring point, and record them as Tmax and Tmin; compare and analyze the temperature of each monitoring point in the area with the set temperature range. When the temperature is greater than the maximum value in the set temperature range, the monitoring point corresponding to the temperature is recorded as a high temperature point; calculate the distance between two adjacent high temperature points and record it as high temperature interval, thereby obtaining several high temperature intervals, and calculate their average to obtain the average of the high temperature intervals and record it as LT; use the set formula The temperature concentration value is calculated, where λ4, λ5, and λ6 are the set weight constants. is the average temperature of each monitoring point in the region; thus, the temperature concentration value TZ i corresponding to any acquisition time in the observation window can be obtained;

[0051] Tunnel depth analysis: Compare and analyze the tunnel depth of each monitoring point with the set depth interval. When the tunnel depth is greater than the maximum value in the set depth interval, the monitoring point corresponding to the tunnel depth is recorded as a depth-exceeding point; when the tunnel depth is within the set depth interval, the monitoring point corresponding to the tunnel depth is recorded as a depth-centering point; when the tunnel depth is less than the minimum value in the set depth interval, the monitoring point corresponding to the tunnel depth is recorded as a depth-upper point; the number of depth-exceeding points, depth-centering points and depth-upper points in the region is counted respectively, and they are recorded as S1, S2, and S3 respectively; when S1≥S2+S3, the tunnel depth corresponding to the depth-exceeding point is averaged as the overall depth of the region; when S3≥S1+S2, the tunnel depth corresponding to the depth-upper point is averaged as the overall depth of the region; in other cases, the tunnel depth of each monitoring point in the region is averaged as the overall depth of the region; thus, the overall depth of the region corresponding to any acquisition time in the observation time window is recorded as Ri;

[0052] It should be noted that as tunnel excavation progresses, the tunnel depths in various areas will change; the tunnel depths in different areas will vary during different stages of excavation;

[0053] Step 3: Real-time monitoring and analysis of microseismic events:

[0054] Obtain the number of microseismic events and the position, amplitude, and energy value corresponding to each microseismic event (the energy of a microseism refers to the energy released during the rupture or sliding of the rock mass under stress when a microseismic event occurs); perform weighted calculation on the amplitude and energy to obtain the microseismic intensity value, and multiply it by the set distance coefficient to obtain the position influence distance; take the position of the microseismic event as the center point, draw a circle with the position influence distance as the radius, record the area within the circle as the influence area, use the centroid algorithm to obtain the center point of each influence area, calculate the distance between the center point and the position of the microseismic event and record it as the microseismic distance of the influence area; divide the microseismic intensity by the microseismic distance to obtain the influence intensity, thereby obtaining the influence intensity of each microseismic event on the influence area, and sum them to obtain the microseismic event cumulative constant of all microseismic events on the influence area, thereby obtaining the microseismic event cumulative constant of each area;

[0055] Step 4: Calculate the stress concentration value YZi, temperature concentration value TZi and overall depth Ri at each acquisition moment in the observation time window through the set formula Bi=k1×YZi+k2×TZi+k3×Ri to obtain the rockburst risk value Bi corresponding to each acquisition moment, where k1, k2, and k3 are set weight constants respectively; construct a two-dimensional rectangular coordinate system with time as the horizontal coordinate and risk value as the vertical coordinate, input each rockburst risk value in the observation time window into the coordinate system in the order of the corresponding acquisition moments, and record the position of the rockburst risk value in the coordinate axis as the risk point, and use a smooth curve to connect each risk point in turn to obtain a curve of the rockburst risk value changing with time; draw a tangent to the curve at each risk point, and use the least squares method to calculate the slope of each tangent and record it as the risk slope Fi; sum the risk slopes greater than zero to obtain the risk increase value recorded as C1, and sum the risk slopes less than zero and take the absolute value to calculate the risk reduction value recorded as C2; use the set formula The risk index FC of the observation time window is calculated, where β1, β2, and β3 are the set weight constants. is the mean value of the risk slope of each risk point, and μ is the cumulative constant of microseismic events. It can be seen from the formula that when the rockburst risk value in the observation window increases faster (that is, the greater the risk increase value, the smaller the risk reduction value), the greater the risk of rockburst in the area, the greater the risk index; when the rockburst risk value in the area is more unstable, the risk index is greater.

[0056] Step 5: Divide the risk index corresponding to the observation time window closest to the current time of the system by the set time window constant (this time window constant is set by the technicians in this field according to the actual situation of the tunnel excavation project) to obtain the latest observation time window, thereby obtaining the latest observation time window of each area, and updating it to step 1; it can be easily known from the calculation method of the observation time window that the greater the rock burst risk of the area, the shorter the next observation time window of the area, and the observation time window can be dynamically adjusted according to the actual rock burst risk status of the area, so as to observe the abnormal situation of the area in time;

[0057] Send the risk index of the observation time window to the safety warning module;

[0058] By integrating multiple parameters such as stress, temperature, tunnel depth, accumulation of microseismic events, etc. for comprehensive analysis, rather than single-dimensional data, the multi-dimensional fusion analysis method improves the accuracy of rock burst risk prediction and can identify potential high-risk areas; at the same time, the observation time window length of multi-dimensional data analysis is dynamically adjusted according to the rock burst risk of each area. When the rock burst risk increases, the observation time window will be shortened, so that abnormal changes in the tunnel can be captured more timely and a faster response can be achieved. This dynamic adjustment mechanism improves the flexibility and response speed of tunnel monitoring; it helps engineers take preventive measures in advance during tunnel excavation, reduces the possibility of sudden rock burst accidents, and provides data support for achieving safety in the process of tunnel excavation.

[0059] The safety warning module plans safe paths for workers in each area based on the rockburst risk value of each area, so as to achieve rapid evacuation under rockburst risk and maximize the safety of tunnel construction workers; specifically:

[0060] The risk index corresponding to each received observation time window is recorded as FCa, where a = 1, 2, 3 ... A, A is a positive integer, A is the total number of observation time windows, and a is the serial number of any observation time window; the observation time windows are numbered in the order of their corresponding time, with the serial number of the observation time window as the horizontal coordinate and the risk index as the vertical coordinate. The risk index is input into the coordinate axis according to its corresponding number to obtain the index point, and the index points are connected in sequence with a smooth curve to obtain the risk index change curve; a tangent line is drawn at each index point, and the slope of the tangent line is calculated using the least squares method to obtain the trend slope of each index point, which is recorded as Pa; the trend slopes greater than zero are summed up to obtain the index increase value, which is recorded as H1, and the trend slopes less than zero are summed up and the absolute value is taken to calculate the index decrease value, which is recorded as H2; using the set formula The rockburst warning value PH is calculated, where β3 and β4 are the set weight constants. is the average trend slope of each index point; the rock burst warning value is compared with the set warning threshold value. When the rock burst warning value of any area is greater than the set warning threshold value, it means that there is a rock burst risk in the area. Then, a safety warning is issued to the tunnel construction personnel, and the area corresponding to the rock burst warning value is recorded as a risk area. Thus, one or more risk areas can be obtained;

[0061] Safety warning: communicate with the smart devices of each construction worker in the tunnel (the smart devices here include but are not limited to smart watches and smart phones) to obtain the location of each worker in real time and mark it on the tunnel map;

[0062] It is assumed that there are several refuge points inside the tunnel and they are marked on the tunnel map. Usually, the location of the refuge point is specially designed and reinforced, with a stronger structure, good ventilation and lighting, and equipped with necessary emergency supplies, and is not easily affected by risks such as vibration and blasting. Usually, there are several refuge points distributed inside the tunnel to deal with emergencies. The location of the construction workers inside the tunnel is taken as the starting point, and the nearest refuge point is selected as the end point. The path planning algorithm (such as Di jkstra algorithm) to obtain several paths leading to the nearest refuge point, and mark each path on the tunnel map to calculate the areas involved in each path and the length of the road section involved; select any one of the paths, record the area involved in the path as the relevant area and its corresponding section length; retrieve the rock burst warning value of the relevant area, normalize it with the section length and take its number, and perform weighted calculation on the value to obtain the section risk value; thereby, the section risk value of each relevant area of ​​the path can be obtained, and the sum of the values ​​can be calculated to obtain the path risk value, thereby obtaining the path risk values ​​of several paths, and select the path with the smallest path risk value as the target path of the construction employee, and send it to the smart device of the construction personnel to guide them to take emergency risk avoidance;

[0063] By fitting a curve to the changes in the rockburst risk index within the observation time window and analyzing it based on the trend slope, the system can dynamically monitor the risk change trend of each area, accurately identify areas with increased risks, and issue warnings in advance to prevent construction workers from entering high-risk areas. When a rockburst warning is triggered, the Dijkstra and other path planning algorithms are used to plan the path from the construction worker's location to the nearest refuge point. By combining the rockburst warning values ​​of different paths, the length of the road section and other factors, a comprehensive assessment of the safety of each path is conducted, and the refuge path with the lowest risk can be selected for each construction worker, significantly improving the efficiency and safety of risk avoidance, so as to achieve rapid refuge under rockburst risk and maximize the safety of tunnel construction workers.

[0064] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. A real-time monitoring system for tunnel construction based on big data, characterized in that: include: Real-time monitoring module and safety early warning module; The real-time monitoring module obtains the risk index within the observation time window based on the fusion analysis of multi-dimensional factors; The risk index is divided by the set time window constant to obtain the latest observation time window, from which the latest observation time window of each region can be obtained. Each region performs multi-dimensional factor fusion analysis according to the latest observation time window; The latest observation time window of the safety warning module is based on the risk index of each area in each observation time window to perform development trend analysis to determine whether to generate a rock burst warning. If a rock burst warning is triggered, a safety warning will be issued to the construction personnel inside the tunnel; Specifically: The received observation time windows are numbered according to their corresponding time sequence, with the sequence number of the observation time window as the horizontal coordinate and the risk index as the vertical coordinate. The risk index is input into the coordinate axis according to its corresponding number to obtain the index point, and a smooth curve is used to connect the index points in sequence to obtain a risk index change curve chart; according to the risk index change curve chart, a development trend analysis is performed to obtain a rock burst warning value; The rockburst warning value is compared with the set warning threshold. When the rockburst warning value of any area is greater than the set warning threshold, a safety warning is issued to the tunnel construction personnel, and the area corresponding to the rockburst warning value is recorded as a risk area. Thus, one or more risk areas can be obtained. When a security warning is triggered, the specific warning steps are as follows: By communicating with the smart devices of each construction worker in the tunnel, the location of each construction worker can be obtained in real time and marked on the tunnel map; It is assumed that there are several refuge points inside the tunnel, and they are marked on the tunnel map; the location of the construction workers inside the tunnel is taken as the starting point, and the nearest refuge point is selected as the end point. The path planning algorithm is used to obtain several paths leading to the nearest refuge point, and each path is marked on the tunnel map to calculate the areas involved in each path and the length of the road section involved; Optional For one of the paths, the area involved in the path is recorded as the relevant area and the corresponding section length; the rock burst warning value of the relevant area is retrieved, and it is normalized with the section length and its number is taken, and the value is weighted and calculated to obtain the section risk value; the section risk value of each relevant area of ​​the path can be obtained, and the sum of them is calculated to obtain the path risk value, thereby obtaining the path risk values ​​of several paths, and the path with the smallest path risk value is selected as the target path of the construction employee, and it is sent to the construction personnel's smart device.

2. A real-time monitoring system for tunnel construction based on big data according to claim 1, characterized in that: The specific steps for developing trend analysis based on the risk index change curve are as follows: Draw a tangent line at each index point, and use the least squares method to calculate the slope of the tangent line to obtain the trend slope of each index point; The trend slopes greater than zero are summed to obtain the exponential increase value, and the trend slopes less than zero are summed and the absolute value is taken to obtain the exponential decrease value; The risk index, trend slope, index increase value and index decrease value of each observation time window are normalized and their values ​​are taken, and the rockburst warning value is obtained by numerical analysis.

3. The real-time monitoring system for tunnel construction based on big data according to claim 1 is characterized in that: The specific steps based on multi-dimensional factor fusion analysis are as follows: 301: Assume that there is an observation time window; divide the tunnel information at each acquisition time in the region by the observation time window segment, thereby obtaining the tunnel information in each observation time window; wherein the tunnel information includes the stress, temperature and tunnel depth of each monitoring point; 302: Perform stress concentration analysis based on the stress of each monitoring point in the region to obtain a stress concentration value, thereby obtaining the stress concentration value at each acquisition time; 303: Based on the temperature of each monitoring point in the area, a stress concentration analysis is performed to obtain the temperature concentration value, thereby obtaining the temperature concentration value at each acquisition time: 304: Perform tunnel depth analysis based on the tunnel depths of each monitoring point in the area to determine the overall depth of each area, thereby obtaining the overall depth at each acquisition time; 305: Performing microseismic event accumulation analysis based on the number of microseismic events to obtain a microseismic event accumulation constant; 306: normalizing the stress concentration value, temperature concentration value and overall depth at each acquisition time within the observation time window and taking their values, and performing numerical analysis to obtain the rock burst risk value corresponding to each acquisition time; A two-dimensional rectangular coordinate system is constructed with time as the horizontal coordinate and risk value as the vertical coordinate. Each rockburst risk value in the observation time window is input into the coordinate system in the order of its corresponding acquisition time, and the position of the rockburst risk value in the coordinate axis is recorded as the risk point. A smooth curve is used to connect each risk point in turn to obtain a curve graph of rockburst risk value changing with time. 307: Draw a tangent line of the curve at each risk point, and calculate the slope of each tangent line using the least square method and record it as the risk slope; sum the risk slopes greater than zero to obtain the risk increase value, and sum the risk slopes less than zero and take the absolute value to obtain the risk reduction value; The cumulative constant of microseismic events, rockburst risk value at each acquisition moment, risk slope, risk increase value and risk reduction value are formulated and analyzed to obtain the risk index of the observation time window.

4. A real-time monitoring system for tunnel construction based on big data according to claim 3, characterized in that: Stress Concentration Analysis: The maximum stress value and the minimum stress value are selected from the stress corresponding to each monitoring point, and the stress of each monitoring point is compared and analyzed with the set stress interval. When the stress is greater than the maximum value in the set stress interval, the monitoring point corresponding to the stress is recorded as a strong stress point, thereby obtaining several strong stress points in the region, and the distance between two adjacent strong stress points is calculated and recorded as a strong stress interval, thereby obtaining several strong stress intervals, and the average value thereof is calculated to obtain the mean value of the strong stress interval in the region; The stress, maximum stress, minimum stress and mean value of strong stress interval at each monitoring point are normalized and their values ​​are taken, and the stress concentration value is obtained through numerical analysis.

5. A real-time monitoring system for tunnel construction based on big data according to claim 4, characterized in that: Temperature concentration analysis: The maximum temperature and the minimum temperature are selected from the temperatures corresponding to each monitoring point, and the temperature of each monitoring point in the area is compared and analyzed with the set temperature range. When the temperature is greater than the maximum value in the set temperature range, the monitoring point corresponding to the temperature is recorded as a high temperature point; the distance between two adjacent high temperature points is calculated and recorded as a high temperature interval, thereby obtaining several high temperature intervals, and the average value of the high temperature intervals is obtained by averaging them; The temperature, maximum temperature, minimum temperature and mean high temperature interval of the monitoring point are normalized and their values ​​are taken, and the temperature concentration value is obtained by numerical analysis.

6. A real-time monitoring system for tunnel construction based on big data according to claim 5, characterized in that: Tunnel Depth Analysis: Compare and analyze the tunnel depth of each monitoring point with the set depth interval. When the tunnel depth is greater than the maximum value in the set depth interval, the monitoring point corresponding to the tunnel depth is recorded as a depth-exceeding point; when the tunnel depth is within the set depth interval, the monitoring point corresponding to the tunnel depth is recorded as a depth-centering point; when the tunnel depth is less than the minimum value in the set depth interval, the monitoring point corresponding to the tunnel depth is recorded as a depth-upper point; the number of depth-exceeding points, depth-centering points, and depth-upper points in the area is counted respectively, and they are recorded as S1, S2, and S3 respectively; When S1≥S2+S3, the tunnel depths corresponding to the depth-exceeding points are averaged as the overall depth of the area; when S3≥S1+S2, the tunnel depths corresponding to the upper depth points are averaged as the overall depth of the area; in other cases, the tunnel depths of each monitoring point in the area are averaged as the overall depth of the area; thus, the overall depth of the area at any acquisition time within the observation time window can be obtained.

7. A real-time monitoring system for tunnel construction based on big data according to claim 6, characterized in that: Cumulative analysis of microseismic events: The number of microseismic events and the position, amplitude and energy value corresponding to each microseismic event are obtained; the amplitude and energy are weighted to obtain the microseismic intensity value, and the value is multiplied by the set distance coefficient to obtain the position influence distance; the position of the microseismic event is taken as the center point, and a circle is drawn with the position influence distance as the radius, and the area within the circle is recorded as the influence area, and the center point of each influence area is obtained by using the centroid algorithm, and the distance between the center point and the position of the microseismic event is calculated and recorded as the microseismic distance of the influence area; the microseismic intensity is divided by the microseismic distance to obtain the influence intensity, thereby obtaining the influence intensity of each microseismic event on the influence area, and the sum is calculated to obtain the microseismic event cumulative constant of all microseismic events on the influence area, thereby obtaining the microseismic event cumulative constant of each area.

Images