Monitoring and early warning method for mine engineering dam body

Abstract

A monitoring and early warning method for a mine engineering dam body comprises the following steps: according to different types and structural characteristics of the mine engineering dam body, installing monitoring equipment at a key position; real-time monitoring data of a dam body are collected through a data collection system, the monitoring data in one year are comprehensively analyzed, abnormal data are intelligently recognized in combination with a machine learning technology, and an early warning threshold value is calculated and set; the strain monitoring of the residual coal pillar dam body is mainly used for monitoring the internal stress change of a main bearing structure; for an artificial dam body constructed by concrete, detecting small deformation of the artificial dam body by adopting fiber bragg grating strain equipment or a surface strain gauge; in combination with machine learning and similar engineering experience, singular data are automatically eliminated, and a monitoring point early warning threshold value with the confidence coefficient being 95% or above is obtained through an algorithm; and real-time monitoring data are imported into a numerical simulation system, so that visualization of stress, strain and seepage states of the whole dam body is realized. According to the method, the monitoring accuracy of different dam bodies is improved, and early warning can be provided for possible structural instability risks.

Description

Technical Field

[0001] The invention relates to a monitoring and early warning method for a mining engineering dam body, and belongs to the technical field of mining engineering safety monitoring. Background Art

[0002] The dam of mining projects includes tailings pond dam, abandoned coal mine dam and underground coal mine reservoir dam. In mining projects, the safety and stability of the dam structure are directly related to the safety of the mining area and the surrounding environment. The dam of mining projects not only has to withstand the long-term effects of rock and water pressure, but also faces the impact caused by mining disturbance. With the increase of mining depth and the expansion of mining scale, the stress distribution and seepage characteristics of the mine dam gradually become more complicated. Especially in the process of deep mining, rock deformation, water seepage and pressure changes often pose a potential threat to the safety of the dam.

[0003] At present, the monitoring technology of mining engineering dams is gradually developing, but conventional monitoring methods mostly focus on the changes of a single parameter or a few parameters, which cannot fully grasp the overall state of the dam, and it is difficult to achieve all-round, multi-parameter accurate monitoring in a complex mining environment. In particular, dams such as tailings ponds and coal mine underground reservoirs face great risks of stability due to long-term multiple influences of water erosion, stress changes and coal seam subsidence. Traditional monitoring and early warning methods have limitations in monitoring sensitivity, comprehensive data analysis, and early warning threshold setting. Usually, threshold setting relies on manual experience or theoretical values, and fails to fully consider the diversity of the on-site geological environment and data volatility, resulting in insufficient early warning accuracy of the monitoring system.

[0004] In addition, different types of mining engineering dams have differences in materials and structures. For example, the coal pillar dam of a coal mine and the artificially constructed concrete dam have different mechanical properties: the coal pillar dam is usually relatively weak and prone to deformation, and is suitable as a supporting structure for the goaf; while the concrete dam is used for water isolation and water retention, and has less deformation. These structural differences put forward higher requirements for monitoring methods, and it is necessary to select appropriate monitoring equipment and methods according to the specific characteristics and uses of the dam. Summary of the invention

[0005] The present invention provides a monitoring and early warning method for a mining engineering dam body, which can improve the monitoring accuracy of different dam bodies, and can also provide early warning for possible structural instability risks, thereby effectively reducing the occurrence of safety accidents.

[0006] In order to achieve the above object, the present invention provides a monitoring and early warning method for a mining engineering dam, comprising the following steps:

[0007] S1. Layout and installation of monitoring equipment: According to the different types and structural characteristics of the mining engineering dam, surface strain gauges, borehole stress gauges, penetration pressure gauges and optical fiber strain gauges are installed at key locations;

[0008] S2. Collection and analysis of monitoring data: The real-time monitoring data of the dam body is collected through the data collection system, and the monitoring data within one year is comprehensively analyzed to identify the historical fluctuation pattern, and the abnormal data is intelligently identified by combining machine learning technology, and the warning threshold is calculated and set; this process can avoid the limitations of manually determining the alarm threshold and improve the reliability and accuracy of the warning through big data analysis;

[0009] S3. Differentiated monitoring: Taking into account the differences in material properties of different mining engineering dams, differentiated monitoring methods are adopted according to the actual use and material properties of the dam. The strain monitoring of the residual coal pillar dam is mainly to monitor the changes in internal stress of the main bearing structure; for artificial dams constructed of concrete, fiber grating strain equipment or surface strain gauges are used to detect their smaller deformations to ensure that different dam materials are accurately monitored;

[0010] S4. Data analysis and intelligent early warning: Combine machine learning with similar engineering experience to automatically eliminate singular data, and use algorithms to derive early warning thresholds for monitoring points with a confidence level of more than 95%. Import real-time monitoring data into the numerical simulation system to visualize the stress, strain, and seepage status of the entire dam body.

[0011] Furthermore, the key positions in S1 are the waterproof and sealed wall, the water-retaining coal pillars, and the water inlet and outlet pipes.

[0012] Furthermore, the comprehensive analysis of the monitoring data within one year in S2 is mainly for verifying the data between the sensors in the coal pillar dam body, the stress evolution trend of the coal pillar dam body and the strain evolution trend of the concrete dam body, and the current monitoring content of the coal pillar dam body and the historical monitoring content.

[0013] Furthermore, the specific process of data analysis and intelligent early warning in S4 is as follows:

[0014] S4.1. Determine the weight value of each monitoring data using the analytic hierarchy process and variable set mathematical theory; determine the membership distribution function value of each monitoring content according to the evolution of each monitoring data, and determine it in the following way:

[0015]

[0016] In the formula, x i is each monitoring data item, w i is the weight value of each monitoring data item, f(x i ) is the membership function value corresponding to the monitoring data, and H is the calculated stability index of the mining engineering dam;

[0017] S4.2. Use indoor experiments and numerical simulation methods to determine the dangerous locations in the monitoring area, and combine the historical content of the monitoring data in the area to comprehensively determine the warning limit of the monitoring content. The maximum value of the monitoring data excluding singular points within one year is used as the upper limit of data monitoring and warning. The upper limit of monitoring and warning is determined by the following formula:

[0018]

[0019] In the formula, j is the number of days from the current monitoring time, It is the monitoring data set after removing data singularity points;

[0020] S4.3. After removing the data singularity points, the monitoring data set is brought into the numerical calculation software for synchronous calculation and full-field analysis of the monitoring data, and the stress, strain, seepage law and stability index distribution of the monitoring field are displayed in real time. The numerical calculation software brought in should be mature commercial geotechnical engineering numerical calculation software, such as Ansys and Abaqus. The frequency of numerical calculation is mainly affected by the performance of the server processor and may vary. The calculation interval is to calculate the full-field data distribution once every 2 hours of monitoring data.

[0021] The present invention installs surface strain gauges, borehole stress gauges, permeability pressure gauges and optical fiber strain gauges at key positions according to different types and structural characteristics of mining engineering dam bodies; collects real-time monitoring data of the dam body through a data acquisition system, and conducts a comprehensive analysis of the monitoring data within one year to identify historical fluctuation patterns, and intelligently identifies abnormal data in combination with machine learning technology, calculates and sets warning thresholds; the strain monitoring of the residual coal pillar dam body is mainly to monitor the internal stress changes of the main bearing structure; for the artificial dam body constructed of concrete, a fiber grating strain device or a surface strain gauge is used to detect its smaller deformation to ensure that different dam body materials are accurately monitored; combined with machine learning and similar engineering experience, singular data is automatically eliminated, and the warning threshold of the monitoring point with a confidence level of more than 95% is obtained through an algorithm; the real-time monitoring data is imported into a numerical simulation system to realize the visualization of the full-field stress, strain and seepage state of the dam body. Compared with the prior art, the present invention improves the monitoring accuracy of different dam bodies, and can also provide early warning for possible structural instability risks, effectively reducing the occurrence of safety accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a work flow chart of the present invention;

[0023] Figure 2 It is a schematic diagram of the monitoring of the concrete dam of a mining project and the deployment of the monitoring equipment in an embodiment of the present invention;

[0024] Figure 3It is a schematic diagram of the coal pillar dam monitoring and monitoring equipment layout of a mining engineering dam in an embodiment of the present invention;

[0025] Figure 4 This is the measurement result diagram of the strain data of the artificial dam body;

[0026] Figure 5 This is a graph showing the full-field strain data results of the coal and rock mass in the dam area monitored by the present invention.

[0027] In the figure: 1. Surface strain gauge, 2. Permeability pressure gauge, 3. Borehole stress gauge, 4. Optical fiber strain gauge, 5. Top and bottom plate displacement gauge. DETAILED DESCRIPTION

[0028] The present invention will be further described below in conjunction with the accompanying drawings.

[0029] like Figure 1 As shown, a monitoring and early warning method for a mining engineering dam body includes the following steps:

[0030] S1. Layout and installation of monitoring equipment: According to the different types and structural characteristics of the mining engineering dam, surface strain gauges 1, borehole stress gauges 3, penetration pressure gauges 2 and optical fiber strain gauges 4 are installed at key locations;

[0031] S2. Collection and analysis of monitoring data: The real-time monitoring data of the dam body is collected through the data collection system, and the monitoring data within one year is comprehensively analyzed to identify the historical fluctuation pattern, and the abnormal data is intelligently identified by combining machine learning technology, and the warning threshold is calculated and set; this process can avoid the limitations of manually determining the alarm threshold and improve the reliability and accuracy of the warning through big data analysis;

[0032] S3. Differentiated monitoring: Taking into account the differences in material properties of different mining engineering dams, differentiated monitoring methods are adopted according to the actual use and material properties of the dam. The strain monitoring of the residual coal pillar dam is mainly to monitor the changes in internal stress of the main bearing structure; for artificial dams constructed of concrete, fiber grating strain equipment or surface strain gauges are used to detect their smaller deformations to ensure that different dam materials are accurately monitored;

[0033] S4. Data analysis and intelligent early warning: Combine machine learning with similar engineering experience to automatically eliminate singular data, and use algorithms to derive early warning thresholds for monitoring points with a confidence level of more than 95%. Import real-time monitoring data into the numerical simulation system to visualize the stress, strain, and seepage status of the entire dam body.

[0034] Example: Figure 2 and Figure 3As shown, monitoring equipment such as surface strain gauges 1, permeability pressure gauges 2, borehole stress gauges 3 and buried optical fiber strain gauges 4 are arranged at key locations such as waterproof closed walls, water-retaining coal pillars, and water inlet and outlet pipes to cover important parameters such as stress, strain and seepage of the dam body. Among them, the surface strain gauge 1 is arranged in an area not less than 300mm away from the top and bottom plates and the left and right side tunnels. It is only arranged on the surface of the mining engineering dam body with concrete as the main material, and is arranged along the four diagonals of the dam body surface; the permeability pressure gauge 2 is arranged at 0.5m away from the bottom plate, with a drill hole diameter of 48-50mm and a hole depth of 4m. It is arranged at the junction of the coal pillar dam body and the concrete dam body, or the junction area between the concrete dam body and other geological bodies; the borehole stress gauge 3 is arranged in the middle area of ​​the coal pillar dam body, and four gauges are evenly arranged horizontally along the coal pillar dam body, with a drill hole diameter of 44-46mm and a hole depth of 3m; the optical fiber strain gauge 4 is buried inside the concrete dam body, arranged in an area not less than 300mm away from the top and bottom plates, and is pre-buried inside the concrete dam body when the concrete dam body is poured; the top and bottom plate displacement meter 5 is mainly used to monitor the deformation between the top and bottom plates;

[0035] The monitoring data is collected and classified, and uploaded to the data processor. The data collection and storage time may vary depending on the performance of the processor and storage device. It is generally recommended to collect data every 20 seconds. After the monitoring data is collected, the monitoring data will be verified with other monitoring contents of this project, such as: data verification between sensors in the coal pillar dam body, verification of stress evolution trend of the coal pillar dam body and strain evolution trend of the concrete dam body, verification of current monitoring content of the coal pillar dam body and historical monitoring content, etc. The monitoring results are as follows: Figure 4 As shown;

[0036] Different monitoring methods are adopted according to the material properties of different mining engineering dam bodies. For coal pillar dam bodies with large deformation, stress gauges are mainly used to monitor the changes in internal stress; for concrete dam bodies with small deformation, surface strain gauges or fiber grating strain devices are used for strain monitoring to ensure accurate monitoring of dam bodies made of different materials. The installation methods of sensors between different dam bodies are as follows: Figure 2 and Figure 3 As shown;

[0037] The monitoring data of other similar projects are introduced to build a large database of mining engineering dams. The machine learning algorithm is introduced to learn the historical monitoring data of various mining engineering dams and eliminate the singular data in the evolution process of the mining dam. In addition, a monitoring and early warning algorithm is constructed to calculate the stability index of the mining engineering dam. Specifically, the weight value of each data can be determined by using the hierarchical analysis method, variable set mathematical theory, etc.; according to the evolution of each data, the membership distribution function value of each monitoring content is determined, and it is determined by the following method:

[0038]

[0039] In the formula, x i is each monitoring data item, w i is the weight value of each monitoring data item, f(x i ) is the membership function value corresponding to the monitoring data, and H is the calculated stability index of the mining engineering dam;

[0040] At the same time, indoor experiments, numerical simulations and other methods are used to determine the dangerous locations in the monitoring area, and combined with the historical content of the monitoring data in the area, the warning limit of the monitoring content is comprehensively determined. This patent recommends statistically calculating the maximum value of the monitoring data within one year excluding singular points as the upper limit of data monitoring warning. The warning limit can be determined by the following formula:

[0041]

[0042] In the formula, j is the number of days from the current monitoring time, It is the monitoring data set after removing data singularity points;

[0043] The monitoring data set after removing the data singularity points is brought into the synchronous numerical calculation software to perform synchronous calculation and full-field analysis of the monitoring data, and real-time display of the stress, strain, seepage law and stability index distribution of the entire monitoring field. Specifically, the monitoring data should be the data set after removing the singularity points, and the numerical calculation software brought in should be mature commercial geotechnical engineering numerical calculation software, such as: Ansys, Abaqus, etc. The frequency of numerical calculation is mainly affected by the performance of the server processor and may vary. It is recommended that the calculation interval is to calculate the full-field data distribution once every 2 hours of monitoring data. The calculation results are as follows Figure 5 As shown in the figure, it can be seen that the present invention can accurately monitor different dam bodies and provide early warning for possible risks of structural instability.

Claims

1. A monitoring and early warning method for a mining engineering dam, characterized in that: The steps include: S1. Layout and installation of monitoring equipment: According to the different types and structural characteristics of the mining engineering dam, surface strain gauges (1), borehole stress gauges (3), penetration pressure gauges (2) and optical fiber strain gauges (4) are installed at key locations; S2. Collection and analysis of monitoring data: The real-time monitoring data of the dam body is collected through the data collection system, and the monitoring data within one year is comprehensively analyzed to identify the historical fluctuation patterns, and the abnormal data is intelligently identified by combining machine learning technology, and the warning threshold is calculated and set; S3. Differentiated monitoring: For the strain monitoring of the residual coal pillar dam, the focus is on monitoring the changes in the internal stress of the main bearing structure; for the artificial dam constructed of concrete, fiber Bragg grating strain equipment or surface strain gauges are used to detect its smaller deformation; S4, Data analysis and intelligent early warning: Combine machine learning with similar engineering experience to automatically remove singular data and use algorithms to derive early warning thresholds for monitoring points with a confidence level of more than 95%; Import the real-time monitoring data into the numerical simulation system to visualize the stress, strain and seepage status of the dam body.

2. The monitoring and early warning method for a mining engineering dam according to claim 1 is characterized in that: The key positions in S1 are the waterproof closed wall, the water retaining coal pillar, and the water inlet and outlet pipes.

3. The monitoring and early warning method for a mining engineering dam according to claim 1 or 2, characterized in that: The comprehensive analysis of the monitoring data within one year in S2 is mainly to verify the data between the sensors in the coal pillar dam body, the stress evolution trend of the coal pillar dam body and the strain evolution trend of the concrete dam body, and the current monitoring content of the coal pillar dam body and the historical monitoring content.

4. The monitoring and early warning method for a mining engineering dam according to claim 3 is characterized in that: The specific process of data analysis and intelligent early warning in S4 is as follows: S4.

1. Determine the weight value of each monitoring data using the analytic hierarchy process and variable set mathematical theory; determine the membership distribution function value of each monitoring content according to the evolution of each monitoring data, and determine it in the following way: In the formula, x i is each monitoring data item, w i is the weight value of each monitoring data item, f(x i ) is the membership function value corresponding to the monitoring data, and H is the calculated stability index of the mining engineering dam; S4.

2. Use indoor experiments and numerical simulation methods to determine the dangerous locations in the monitoring area, and combine the historical content of the monitoring data in the area to comprehensively determine the warning limit of the monitoring content. The maximum value of the monitoring data excluding singular points within one year is used as the upper limit of data monitoring and warning. The upper limit of monitoring and warning is determined by the following formula: In the formula, j is the number of days from the current monitoring time, It is the monitoring data set after removing data singularity points; S4.

3. The monitoring data set after eliminating the data singularity points is brought into the numerical calculation software for synchronous calculation and full-field analysis of the monitoring data, and the stress, strain, seepage law and stability index distribution of the whole monitoring field are displayed in real time.

Images