Tunnel monitoring method and system based on vibrating wire sensor

Abstract

The invention discloses a tunnel monitoring method and system based on a vibrating wire sensor, and relates to the technical field of tunnel engineering construction safety monitoring, and the method comprises the steps: collecting the vibration frequency data and spatial position coordinates of the vibrating wire sensor in a monitoring region; the method comprises the following steps: calculating a time difference and an energy attenuation ratio between adjacent sensors by extracting spectrum energy density distribution, and constructing a stress wave propagation field; dividing grids in the propagation field to calculate node stress values, generating a stress field distribution cloud picture and extracting a stress concentration area; calculating a gravity center position coordinate change rate and a stress time sequence curve of the stress concentration area, and determining fracture extension parameters; and calculating a surrounding rock bearing capacity attenuation curve based on the parameters, and outputting an early warning signal when the slope exceeds a threshold value. According to the invention, real-time monitoring and early warning of the tunnel surrounding rock instability risk are realized.

Description

Technical Field

[0001] This invention relates to the field of tunnel construction safety monitoring technology, specifically to a tunnel monitoring method and system based on a vibrating wire sensor. Background Technology

[0002] Currently, commonly used methods for monitoring the stability of surrounding rock mainly include displacement monitoring, stress monitoring, and acoustic emission monitoring. Among them, displacement monitoring mainly assesses the deformation state of the surrounding rock by measuring the axial displacement of anchor bolts and anchor cables; stress monitoring mainly measures the stress changes inside the surrounding rock using stress gauges; and acoustic emission monitoring determines the damage state of the surrounding rock by detecting the acoustic signals generated by the fracturing of the surrounding rock.

[0003] Traditional methods for monitoring surrounding rock have the following problems: a single monitoring method cannot comprehensively reflect the stability state of the surrounding rock. Displacement monitoring can only reflect the deformation characteristics of the surface surrounding rock and cannot reflect the damage evolution process of the deep surrounding rock; although stress monitoring can obtain the internal stress state of the surrounding rock, the distribution of measuring points is limited, making it difficult to reflect the overall distribution characteristics of the stress field; although acoustic emission monitoring can detect the development process of micro-fractures in the surrounding rock, it is greatly affected by environmental noise and it is difficult to accurately locate the fractures.

[0004] Existing monitoring methods generally suffer from delayed early warning. Since rock instability is often a gradual process, relying solely on threshold values ​​for physical quantities such as displacement and stress is insufficient to detect early signs of instability in a timely manner. Furthermore, the lack of quantitative characterization of the rock damage evolution process makes it difficult to accurately assess the decline trend of the rock's bearing capacity, resulting in untimely and inaccurate early warnings.

[0005] Existing monitoring methods rely on relatively simple data analysis techniques, failing to fully extract the surrounding rock damage information contained within the monitoring data. Most methods focus only on the time-domain characteristics of the monitoring data, neglecting the surrounding rock damage mechanisms reflected in the frequency-domain characteristics, making it difficult to achieve refined identification and early warning of the surrounding rock instability process. Summary of the Invention

[0006] The purpose of this invention is to provide a tunnel monitoring method and system based on a vibrating wire sensor, aiming to solve at least one of the technical problems existing in the prior art.

[0007] The technical solution of this invention is: a tunnel monitoring method based on a vibrating wire sensor, comprising the following steps: Collect vibration frequency data and spatial location coordinates of the vibrating wire sensor within the monitoring area; Based on vibration frequency data, the spectral energy density distribution is extracted, the time difference and energy attenuation ratio of the spectral energy density distribution between adjacent vibrating wire sensors are calculated, and the stress wave propagation field of the monitoring area is constructed by combining spatial location coordinates. Within the stress wave propagation field of the monitoring area, a stress calculation grid is divided. The stress values ​​of the grid nodes are calculated based on the spectral energy density distribution and energy attenuation ratio. A stress field distribution cloud map is generated, and the region with the maximum stress gradient in the stress field distribution cloud map is extracted as the stress concentration region. Calculate the stress value time series curve within the stress concentration region, extract the centroid position coordinates of the stress concentration region, and determine the crack propagation parameters based on the time change rate of the centroid position coordinates and the stress value time series curve. The bearing capacity attenuation curve of the surrounding rock is calculated based on the crack propagation parameters. The bearing capacity attenuation threshold of the surrounding rock is determined according to the stress value in the stress concentration area. When the slope of the bearing capacity attenuation curve of the surrounding rock exceeds the bearing capacity attenuation threshold of the surrounding rock, an early warning signal is output.

[0008] Based on vibration frequency data, the spectral energy density distribution is extracted, and the time difference and energy attenuation ratio of the spectral energy density distribution between adjacent vibrating wire sensors are calculated. Combined with spatial location coordinates, the stress wave propagation field of the monitoring area is constructed, including: The vibration frequency data is segmented according to the sampling interval, and the segmented vibration frequency data is processed by a window function to obtain the window function processed signal. The power spectral density is obtained by performing spectral analysis on the signal processed by the window function, and the power spectral density is normalized to obtain the spectral energy density distribution. The cross-correlation coefficient between adjacent vibrating wire sensors is calculated based on the spectral energy density distribution. The time shift corresponding to the maximum value of the cross-correlation coefficient is extracted as the spectral energy density distribution time difference, and the amplitude ratio of the maximum value is extracted as the energy attenuation ratio. The wave velocity distribution is obtained by mapping the spatial location coordinates and the time difference of the spectral energy density distribution to the monitoring area. The wave velocity distribution is then corrected by combining the energy attenuation ratio to generate the stress wave propagation field in the monitoring area.

[0009] Within the stress wave propagation field of the monitoring area, a stress calculation grid is divided. The stress values ​​at the grid nodes are calculated based on the spectral energy density distribution and energy attenuation ratio. A stress field distribution cloud map is generated, and the region with the maximum stress gradient in the stress field distribution cloud map is extracted as the stress concentration region, including: A grid coordinate system is established within the stress wave propagation field of the monitoring area. The stress calculation grid is divided according to the distribution range of the vibrating wire sensor, and the spatial coordinates of the grid nodes in the stress calculation grid are calibrated. The spectral energy density distribution is allocated based on the spatial coordinates of the grid nodes to obtain the grid node energy distribution. The attenuation coefficient of the grid node energy distribution is calculated based on the energy attenuation ratio. The stress values ​​of the grid nodes are calculated using the energy distribution and attenuation coefficient of the grid nodes, and a stress field distribution cloud map is generated based on the stress values ​​of the grid nodes. The stress value change of adjacent grid nodes in the stress field distribution cloud map is calculated, and the region where the stress value change is greater than the change of the surrounding grid nodes is determined as the region with the maximum stress gradient, which is used as the stress concentration region.

[0010] The spectral energy density distribution is allocated based on the spatial coordinates of the grid nodes to obtain the grid node energy distribution. The attenuation coefficient of the grid node energy distribution is calculated based on the energy attenuation ratio, including: Calculate the spatial coordinate distance matrix between the grid nodes and the vibrating wire sensor, divide the spectral energy density distribution into components based on the spatial coordinate distance matrix, and determine the propagation path of the spectral energy density distribution at the spatial coordinates of the grid nodes; Obtain the angle between the propagation path and the normal direction of the tunnel wall, calculate the energy propagation loss based on the angle, correct the energy propagation loss and the spectral energy density distribution, and superimpose the corrected spectral energy density distribution at the spatial coordinates of the grid node to obtain the grid node energy distribution; Calculate the propagation distance between adjacent grid nodes, obtain the propagation attenuation factor based on the propagation distance, and use the propagation attenuation factor to correct the energy attenuation ratio to obtain the grid node energy distribution attenuation coefficient.

[0011] Calculating the stress value time series curve within the stress concentration region, extracting the centroid coordinates of the stress concentration region, and determining the crack propagation parameters based on the time change rate of the centroid coordinates and the stress value time series curve include: Obtain the stress values ​​and coordinates of grid nodes within the stress concentration area, and construct a stress value time series curve by sorting the grid node stress values ​​according to the monitoring time. The stress distribution weight of the stress concentration region is calculated based on the stress values ​​of the grid nodes, and the centroid coordinates of the stress concentration region are calculated based on the grid node position coordinates and the stress distribution weights. Record the change data of the center of gravity position coordinates with the monitoring time, and calculate the displacement of the center of gravity position coordinates within adjacent monitoring time intervals to obtain the time change rate of the center of gravity position coordinates; The stress value time series curve and the rate of change of the centroid position coordinate over time are correlated with the monitoring time, and the change in stress-displacement relationship is calculated as a crack propagation parameter.

[0012] The stress-displacement relationship between the stress-time series curve and the rate of change of the centroid position coordinates over time is calculated according to the monitoring time and used as a crack propagation parameter, including: The stress value time series curve is denoised and normalized to obtain a standardized stress curve. The stress fluctuation frequency and stress fluctuation amplitude are calculated based on the standardized stress curve. The stress fluctuation frequency and stress fluctuation amplitude are combined to construct a stress displacement correspondence matrix. The displacement direction vector is calculated based on the time change rate of the center of gravity position coordinates. The displacement direction vector is mapped and matched with the stress-displacement corresponding matrix. The angle between the main direction and the main direction in the displacement direction vector and the stress-displacement corresponding matrix is ​​extracted. The displacement change intensity is calculated based on the angle between the main direction. The stress-displacement correspondence matrix is ​​corrected using the intensity of displacement change, and the stress-displacement gradient change is calculated based on the corrected stress-displacement correspondence matrix. The stress-displacement gradient change and the displacement change intensity are combined and calculated iteratively to obtain the stress-displacement relationship change, which is then used as the crack propagation parameter.

[0013] The bearing capacity attenuation curve of the surrounding rock is calculated based on the fracture propagation parameters. The bearing capacity attenuation threshold of the surrounding rock is determined according to the stress value in the stress concentration area. When the slope of the bearing capacity attenuation curve of the surrounding rock exceeds the bearing capacity attenuation threshold of the surrounding rock, an early warning signal is output, including: Acquire temporal variation data of fracture propagation parameters, calculate the bearing capacity variation trend within the monitoring period based on the temporal variation data, and generate a bearing capacity attenuation curve of surrounding rock. Collect stress value monitoring data in the stress concentration area, calculate the change range of the stress value monitoring data, and determine the bearing capacity attenuation threshold of the surrounding rock based on the change range of the stress value; The rate of change of the surrounding rock bearing capacity at adjacent monitoring times is calculated based on the bearing capacity decay curve of the surrounding rock and used as the slope of the bearing capacity decay curve. The slope of the bearing capacity attenuation curve of the surrounding rock is compared with the bearing capacity attenuation threshold of the surrounding rock. When the slope of the bearing capacity attenuation curve of the surrounding rock exceeds the bearing capacity attenuation threshold of the surrounding rock, an early warning signal is output.

[0014] This invention provides a tunnel monitoring system based on a vibrating wire sensor, the system comprising: The data acquisition module is used to collect vibration frequency data and spatial position coordinates of the vibrating wire sensor within the monitoring area; The stress wave field construction module is used to extract the spectral energy density distribution based on vibration frequency data, calculate the time difference and energy attenuation ratio of the spectral energy density distribution between adjacent vibrating wire sensors, and construct the stress wave propagation field of the monitoring area by combining spatial location coordinates. The stress field analysis module is used to divide the stress calculation grid within the stress wave propagation field of the monitoring area, calculate the stress value of the grid node based on the spectral energy density distribution and energy attenuation ratio, generate a stress field distribution cloud map, and extract the region with the maximum stress gradient in the stress field distribution cloud map as the stress concentration region. The crack parameter calculation module is used to calculate the stress value time series curve in the stress concentration area, extract the centroid position coordinates of the stress concentration area, and determine the crack propagation parameters based on the time change rate of the centroid position coordinates and the stress value time series curve. The early warning judgment module is used to calculate the bearing capacity attenuation curve of the surrounding rock based on the crack propagation parameters, determine the bearing capacity attenuation threshold of the surrounding rock according to the stress value in the stress concentration area, and output an early warning signal when the slope of the bearing capacity attenuation curve of the surrounding rock exceeds the bearing capacity attenuation threshold of the surrounding rock.

[0015] One technical solution provided in this embodiment of the invention is an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in any of the aforementioned methods.

[0016] One technical solution provided in this embodiment of the invention is a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the steps in any of the aforementioned methods.

[0017] This invention achieves comprehensive capture of stress wave propagation characteristics within a monitored area using a vibrating wire sensor network. The stress wave propagation field constructed based on the spectral energy density distribution accurately reflects the stress distribution state within the surrounding rock. Precise positioning of stress concentration areas is achieved through grid division and stress field cloud mapping, effectively avoiding the limitations of traditional single-point monitoring methods. By analyzing the changes in the centroid coordinates and stress time series characteristics of stress concentration areas, a quantitative relationship between surrounding rock fracture propagation parameters and bearing capacity decay is established, enabling accurate assessment of surrounding rock instability risk. An early warning mechanism based on the slope of the bearing capacity decay curve can promptly detect precursors to surrounding rock instability, significantly improving the timeliness and accuracy of early warnings. This method achieves dynamic monitoring of the surrounding rock stress field distribution, precise identification of fracture propagation processes, and early warning of instability risks, providing reliable technical support for ensuring tunnel construction safety. Attached Figure Description

[0018] Figure 1 A flowchart illustrating a tunnel monitoring method based on a vibrating wire sensor, provided as an embodiment of the present invention; Figure 2 This is a flowchart illustrating the stress concentration region determination process based on grid energy distribution, according to an embodiment of the present invention. Figure 3 This is a schematic diagram of the monitoring system according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention; Figure 5This is a schematic diagram of the storage medium according to an embodiment of the present invention. Detailed Implementation

[0019] like Figure 1 As shown, Figure 1 A flowchart of a tunnel monitoring method based on a vibrating wire sensor provided in this embodiment of the invention, the method comprising the following steps: Collect vibration frequency data and spatial location coordinates of the vibrating wire sensor within the monitoring area; Based on vibration frequency data, the spectral energy density distribution is extracted, the time difference and energy attenuation ratio of the spectral energy density distribution between adjacent vibrating wire sensors are calculated, and the stress wave propagation field of the monitoring area is constructed by combining spatial location coordinates. Within the stress wave propagation field of the monitoring area, a stress calculation grid is divided. The stress values ​​of the grid nodes are calculated based on the spectral energy density distribution and energy attenuation ratio. A stress field distribution cloud map is generated, and the region with the maximum stress gradient in the stress field distribution cloud map is extracted as the stress concentration region. Calculate the stress value time series curve within the stress concentration region, extract the centroid position coordinates of the stress concentration region, and determine the crack propagation parameters based on the time change rate of the centroid position coordinates and the stress value time series curve. The bearing capacity attenuation curve of the surrounding rock is calculated based on the crack propagation parameters. The bearing capacity attenuation threshold of the surrounding rock is determined according to the stress value in the stress concentration area. When the slope of the bearing capacity attenuation curve of the surrounding rock exceeds the bearing capacity attenuation threshold of the surrounding rock, an early warning signal is output.

[0020] Based on vibration frequency data, the spectral energy density distribution is extracted, and the time difference and energy attenuation ratio of the spectral energy density distribution between adjacent vibrating wire sensors are calculated. Combined with spatial location coordinates, the stress wave propagation field of the monitoring area is constructed, including: The vibration frequency data is segmented according to the sampling interval, and the segmented vibration frequency data is processed by a window function to obtain the window function processed signal. The power spectral density is obtained by performing spectral analysis on the signal processed by the window function, and the power spectral density is normalized to obtain the spectral energy density distribution. The cross-correlation coefficient between adjacent vibrating wire sensors is calculated based on the spectral energy density distribution. The time shift corresponding to the maximum value of the cross-correlation coefficient is extracted as the spectral energy density distribution time difference, and the amplitude ratio of the maximum value is extracted as the energy attenuation ratio. The wave velocity distribution is obtained by mapping the spatial location coordinates and the time difference of the spectral energy density distribution to the monitoring area. The wave velocity distribution is then corrected by combining the energy attenuation ratio to generate the stress wave propagation field in the monitoring area.

[0021] A high-precision data acquisition device is used to connect each vibrating wire sensor to collect vibration frequency data. The data acquisition device is equipped with an analog-to-digital converter of 16 bits or more, with a sampling rate set to 1000Hz and a duration of 10 seconds per acquisition, forming a time series containing 10,000 data points. The spatial position coordinates of the vibrating wire sensors are obtained by measuring with a total station, with a measurement accuracy better than ±2mm. The measurement results are stored in three-dimensional coordinate form, with the coordinate origin set at a fixed reference point in the monitoring area.

[0022] After data acquisition, the vibration frequency data was segmented along the time axis. The segment length was set to 1024 data points, with a 50% overlap between adjacent segments (512 data points between each segment). A Hanning window function was applied to each segment to smooth the original signal and reduce spectral leakage. The window function gradually reduces the signal to zero at the edges, thus minimizing frequency leakage in spectral analysis. The signal processed by the window function retains the main frequency characteristics of the original vibration frequency data while improving the accuracy of subsequent spectral analysis.

[0023] The power spectral density is obtained by performing a Fast Fourier Transform on the signal processed by the window function. Power spectral density characterizes the energy distribution of the signal across its frequency components. To facilitate comparison of data from different sensors, the power spectral density is normalized by standardizing the maximum value to 1 and scaling other values ​​proportionally, resulting in the spectral energy density distribution. Normalization eliminates the influence of differences in sensor sensitivity and acquisition gain on the comparison results, making the spectral characteristics of sensors at different locations comparable.

[0024] Based on the obtained spectral energy density distribution, the cross-correlation coefficients between adjacent vibrating wire sensors are calculated. The cross-correlation calculation involves sliding the spectral energy density distribution of one sensor, multiplying it point-by-point with the spectral energy density distribution of the other sensor, and then summing the results. The sliding step size is set to 0.5 mm, and the maximum sliding range is ±100 mm. The maximum value and its corresponding time shift are extracted from the cross-correlation coefficient sequence. This time shift represents the time difference of the spectral energy density distribution, indicating the time required for the stress wave to propagate from one sensor to another. Simultaneously, the amplitude value of the maximum cross-correlation coefficient is extracted, and the ratio of the maximum cross-correlation coefficients of the two sensors is calculated as the energy attenuation ratio. This ratio reflects the energy loss of the stress wave during propagation.

[0025] By combining the spatial coordinates of the sensors with the time difference of the spectral energy density distribution, the stress wave propagation velocity between adjacent sensors is calculated. Based on the propagation velocity data of all adjacent sensor pairs, a wave velocity distribution map within the monitoring area is generated using the Kriging interpolation method. The interpolation grid size is set to 1m × 1m, and the interpolation radius is 1.5 times the maximum distance between the sensors. Considering the energy attenuation of stress waves during propagation, the wave velocity distribution is corrected using the energy attenuation ratio. The correction method adjusts the wave velocity value according to the energy attenuation ratio; the smaller the attenuation ratio, the greater the energy loss, and the more likely there may be abnormalities in the material's mechanical properties, requiring a reduction in the wave velocity value at the corresponding location. The correction coefficient is set according to the logarithmic change of the attenuation ratio: when the attenuation ratio < 0.5, the wave velocity correction coefficient is 0.8; when the attenuation ratio is between 0.5 and 0.8, the correction coefficient changes linearly from 0.8 to 0.9; when the attenuation ratio > 0.8, the correction coefficient ranges from 0.9 to 1.0. Finally, the stress wave propagation field of the monitoring area is generated.

[0026] Using the above method, vibration frequency data collected by vibrating wire sensors was used to construct the stress wave propagation field of the monitoring area. A dataset containing 20 sensors was processed, with an acquisition cycle of once per hour, taking approximately 30 seconds to process. The sensor installation spacing is typically controlled within the range of 5 to 20 meters, with the specific spacing depending on the size and structural complexity of the monitoring area. When stress anomalies occur in the monitoring area, the stress wave propagation velocity in the corresponding area will change significantly. A velocity reduction exceeding 15% should raise concern, and the energy attenuation ratio will also decrease accordingly. Areas with a ratio below 0.6 typically require close monitoring.

[0027] This invention constructs a complete stress wave propagation field by extracting spectral energy density distribution characteristics and combining time difference and energy attenuation ratio, enabling a comprehensive assessment of the stress state in the monitored area. Compared to traditional single-point monitoring methods, this technology provides a more intuitive visualization of stress distribution, making the identification of abnormal stress areas more accurate. The dual indicators of wave velocity distribution and energy attenuation ratio enhance the reliability of monitoring results and reduce the false alarm rate. This method is suitable for stress state monitoring of large structures such as tunnels and underground engineering projects, enabling timely detection of potential risk areas and providing an effective technical means for engineering safety assessment and early warning.

[0028] like Figure 2 As shown, a stress calculation grid is divided within the stress wave propagation field of the monitoring area. The stress values ​​at the grid nodes are calculated based on the spectral energy density distribution and energy attenuation ratio, generating a stress field distribution cloud map. The region with the maximum stress gradient in the stress field distribution cloud map is extracted as the stress concentration region, including: A grid coordinate system is established within the stress wave propagation field of the monitoring area. The stress calculation grid is divided according to the distribution range of the vibrating wire sensor, and the spatial coordinates of the grid nodes in the stress calculation grid are calibrated. The spectral energy density distribution is allocated based on the spatial coordinates of the grid nodes to obtain the grid node energy distribution. The attenuation coefficient of the grid node energy distribution is calculated based on the energy attenuation ratio. The stress values ​​of the grid nodes are calculated using the energy distribution and attenuation coefficient of the grid nodes, and a stress field distribution cloud map is generated based on the stress values ​​of the grid nodes. The stress value change of adjacent grid nodes in the stress field distribution cloud map is calculated, and the region where the stress value change is greater than the change of the surrounding grid nodes is determined as the region with the maximum stress gradient, which is used as the stress concentration region.

[0029] Based on the established stress wave propagation field of the monitoring area, a standardized grid coordinate system is established within the monitoring area. The boundary of the stress calculation grid is determined according to the distribution range of the vibrating wire sensors. The grid boundary is slightly larger than the sensor distribution range to ensure that the calculation range completely covers the monitoring target area. The stress calculation grid adopts an equidistant division method with a grid spacing of 0.5m. This spacing value can be adjusted within the range of 0.1m to 1m according to actual needs; the spacing can be appropriately reduced for areas with high monitoring accuracy requirements. The spatial coordinates of the grid nodes are determined through the correspondence between the grid index and the actual spatial location. Each node is assigned a unique three-dimensional coordinate value, forming a complete grid coordinate system. The total number of grid nodes depends on the size of the monitoring area. For a typical tunnel monitoring area of ​​100m long, 20m wide, and 10m high, approximately 80,000 grid nodes are generated at a 0.5m spacing.

[0030] Energy distribution is performed based on the spatial coordinates of grid nodes using spectral energy density distribution data. An inverse distance weighted interpolation method is employed for energy distribution. For each grid node, its spatial distance to each vibrating wire sensor is calculated, with closer nodes receiving higher weights. The weighting function is designed as the reciprocal of the square of the distance; when the distance is <0.1m, the weight is set to a constant value of 10 to avoid singularities. The energy contribution of each sensor to the current grid node is calculated based on the weights, and the weighted sum is used to obtain the energy distribution value of that node. For areas outside the grid boundaries, an energy attenuation extrapolation method is used, with the extrapolation distance not exceeding 1.5 times the distance between the nearest sensors. Considering energy attenuation during propagation, an energy attenuation ratio is introduced to correct the energy distribution of grid nodes. The attenuation coefficient is calculated based on the energy attenuation ratio between adjacent sensors, using bilinear interpolation to obtain the local attenuation coefficient at the grid node. The attenuation coefficient ranges from 0.6 to 1.0. When the attenuation ratio >0.9, the attenuation coefficient is close to 1.0; when the attenuation ratio drops below 0.5, the attenuation coefficient drops to 0.6, indicating severe energy loss in the area and potential stress anomalies.

[0031] The stress values ​​of the grid nodes are calculated using the energy distribution and attenuation coefficient of the grid nodes. A modified energy-stress conversion method is employed for stress value calculation, which establishes the correlation between spectral energy density and the actual stress field. The calculation process considers energy distribution, attenuation coefficient, and material property parameters, including elastic modulus and Poisson's ratio. The elastic modulus ranges from 20 GPa to 50 GPa, and the Poisson's ratio ranges from 0.2 to 0.3, determined according to the actual engineering geological conditions. For special geological areas, adaptive adjustments can be made by adding correction factors. The calculated grid node stress values ​​cover all grid nodes within the monitoring area. A stress field distribution cloud map is generated based on the grid node stress values. The cloud map is drawn using a cubic spline interpolation algorithm to ensure smooth image transitions and avoid jagged edges. The stress cloud map update frequency is consistent with the data acquisition frequency, typically once per hour, but can be increased to once every 10 minutes for key monitoring areas.

[0032] The stress gradient is calculated as the change in stress value between adjacent grid nodes in the stress field distribution cloud map. The stress gradient is calculated using the central difference method. For internal grid nodes, the average of the absolute values ​​of the stress differences between their four surrounding adjacent nodes is taken as the stress gradient value for that node; for boundary nodes, forward or backward difference is used instead. The stress gradient calculation results form a gradient field distribution, also represented by a pseudo-color cloud map, with high gradient regions indicated by warm tones. To identify stress concentration areas, a stress gradient threshold is set, which is twice the average stress gradient value of the entire region. This parameter can be adjusted within the range of 1.5 to 3 times according to actual monitoring needs. Adjacent grid nodes with stress gradient values ​​greater than the threshold are connected to form regions; these regions are the areas of maximum stress gradient and are marked as stress concentration areas. Stress concentration areas are highlighted on the cloud map with special markers for easy identification by monitoring personnel. Isolated high gradient points with an area less than 5 grid cells are considered noise interference and are not included in the stress concentration areas.

[0033] For example, 32 vibrating wire sensors were deployed in a monitoring area of ​​a tunnel, covering an area of ​​approximately 2000m². 2 The stress calculation grid spacing was set to 0.5m, generating approximately 8000 effective grid nodes. The inverse distance weighting power parameter was set to 2.0 for energy allocation, and the attenuation coefficient correction range was 0.65 to 0.98. The elastic modulus was set to 35 GPa and the Poisson's ratio to 0.25 during the stress-energy conversion process. The calculation results show that the stress values ​​in most areas are in the range of 2 MPa to 5 MPa, with the maximum stress value reaching 8.2 MPa in the stress concentration area. The stress gradient threshold was set to 0.5 MPa / m, and three main stress concentration areas were identified, each with an area of ​​15.5 m². 2 8.7m 2 and 6.2m 2The area of ​​greatest stress concentration was located on the right side of the tunnel arch, which closely matched the crack development area observed on-site, verifying the effectiveness of the method. For low-confidence areas at the edge of the monitoring area, the weight was reduced by increasing the attenuation coefficient, thus minimizing the impact of boundary effects on the results.

[0034] This invention achieves the conversion from discrete sensor data to a continuous stress field distribution by dividing the stress wave propagation field within the monitoring area into a stress calculation grid and combining the spectral energy density distribution and energy attenuation ratio. The generated stress field distribution cloud map intuitively displays the regional stress state, and stress gradient analysis effectively identifies stress concentration areas, providing a quantitative basis for engineering safety assessment. Compared with traditional point monitoring methods, it can comprehensively reflect the stress distribution of the entire monitoring area. Early identification of stress concentration areas can help prevent engineering disasters and improve safety early warning capabilities.

[0035] The spectral energy density distribution is allocated based on the spatial coordinates of the grid nodes to obtain the grid node energy distribution. The attenuation coefficient of the grid node energy distribution is calculated based on the energy attenuation ratio, including: Calculate the spatial coordinate distance matrix between the grid nodes and the vibrating wire sensor, divide the spectral energy density distribution into components based on the spatial coordinate distance matrix, and determine the propagation path of the spectral energy density distribution at the spatial coordinates of the grid nodes; Obtain the angle between the propagation path and the normal direction of the tunnel wall, calculate the energy propagation loss based on the angle, correct the energy propagation loss and the spectral energy density distribution, and superimpose the corrected spectral energy density distribution at the spatial coordinates of the grid node to obtain the grid node energy distribution; Calculate the propagation distance between adjacent grid nodes, obtain the propagation attenuation factor based on the propagation distance, and use the propagation attenuation factor to correct the energy attenuation ratio to obtain the grid node energy distribution attenuation coefficient.

[0036] Calculating the spatial coordinate distance matrix between grid nodes and vibrating wire sensors is fundamental to energy distribution. Assume five vibrating wire sensors are deployed within a monitoring area, each with known coordinates, and the monitoring area is divided into a 20×20×10 grid. For each grid node, the Euclidean distance between it and the five sensors is calculated, forming a 5-dimensional distance vector. Combining the distance vectors of all grid nodes constitutes the spatial coordinate distance matrix between the grid nodes and the vibrating wire sensors.

[0037] The spectral energy density distribution is divided into components based on the spatial coordinate distance matrix. The raw signal acquired by the vibrating wire sensor undergoes a Fourier transform to obtain the spectral energy density distribution, which includes the energy levels of different frequency components. For an energy value at a specific frequency, it is divided into multiple components according to the spatial coordinate distance matrix, with each component corresponding to a propagation path from the sensor to the grid node. For example, if the sensor acquires an energy value of 0.8J at a frequency of 100Hz, this 0.8J of energy may be distributed across multiple propagation paths in different proportions according to the spatial coordinate distance matrix.

[0038] Based on the shortest path principle, there exists an optimal energy propagation path from the sensor to each grid node. The optimal propagation path is calculated using Dijkstra's algorithm; this path is typically the shortest distance or the path with the least energy loss. If the sensor's location coordinates are (5, 8, 2) and a grid node's location coordinates are (12, 15, 5), the calculated propagation path might be a series of intermediate point coordinates, representing the path of energy propagation from the sensor to that grid node.

[0039] Obtain the angle between the propagation path and the normal direction of the tunnel wall, and calculate the energy propagation loss based on this angle. The tunnel wall has a normal direction at each point, representing the direction perpendicular to the wall. When energy propagates along the path, there is an angle between the path direction and the wall normal direction. The larger this angle, the greater the energy propagation loss; the smaller the angle, the smaller the energy propagation loss. Assuming the direction vector of a propagation path is (0.6, 0.8, 0), and the normal direction of the tunnel wall at that point is (0, 1, 0), the calculated angle is approximately 36.9 degrees, corresponding to an energy propagation loss coefficient of 0.2.

[0040] The energy propagation loss and spectral energy density distribution are corrected. For each energy component along the propagation path, the corresponding energy propagation loss coefficient is multiplied to obtain the corrected energy value. For example, if the spectral energy density distribution component along a certain propagation path is 0.3 J, and the corresponding energy propagation loss coefficient is 0.2, then the corrected energy component is 0.24 J. All corrected spectral energy density distribution components are superimposed at the spatial coordinates of the grid node to obtain the energy distribution value of that grid node. If a grid node receives corrected energy components of 0.24 J, 0.18 J, and 0.32 J from three sensors, respectively, then the energy distribution value of that grid node is 0.74 J.

[0041] Calculate the propagation distance between adjacent grid nodes. For any two adjacent grid nodes, calculate the Euclidean distance between them. For example, if the spatial coordinates of adjacent grid nodes are (10, 12, 5) and (10, 13, 5), then the propagation distance between them is 1 unit. Obtain the propagation attenuation factor based on the propagation distance; the larger the propagation distance, the larger the attenuation factor; the smaller the propagation distance, the smaller the attenuation factor. Assuming a propagation distance of 1 unit, the corresponding propagation attenuation factor is 0.9.

[0042] The energy attenuation ratio represents the degree of energy attenuation during propagation and is related to the medium properties and propagation environment. The initial energy attenuation ratio is obtained through actual measurement or theoretical calculation, and then corrected by a propagation attenuation factor to obtain the grid node energy distribution attenuation coefficient. For example, if the initial energy attenuation ratio is 0.85 and the propagation attenuation factor is 0.9, then the corrected grid node energy distribution attenuation coefficient is 0.765.

[0043] This invention achieves precise allocation of spectral energy density distribution and accurate calculation of the energy distribution attenuation coefficient of grid nodes through the aforementioned method. It overcomes the shortcomings of traditional methods that neglect the influence of the angle between the propagation path and the tunnel wall normal direction during energy allocation, thus improving the accuracy of energy allocation. This method fully considers the complexity of energy propagation in the tunnel environment, and by introducing factors such as propagation path, angle calculation, and energy propagation loss, it achieves precise monitoring and assessment of the health status of the tunnel structure. Through the analysis of the energy distribution and attenuation coefficient of grid nodes, abnormal areas in the tunnel structure can be accurately identified, providing reliable technical support for tunnel safety early warning and maintenance.

[0044] Calculating the stress value time series curve within the stress concentration region, extracting the centroid coordinates of the stress concentration region, and determining the crack propagation parameters based on the time change rate of the centroid coordinates and the stress value time series curve include: Obtain the stress values ​​and coordinates of grid nodes within the stress concentration area, and construct a stress value time series curve by sorting the grid node stress values ​​according to the monitoring time. The stress distribution weight of the stress concentration region is calculated based on the stress values ​​of the grid nodes, and the centroid coordinates of the stress concentration region are calculated based on the grid node position coordinates and the stress distribution weights. Record the change data of the center of gravity position coordinates with the monitoring time, and calculate the displacement of the center of gravity position coordinates within adjacent monitoring time intervals to obtain the time change rate of the center of gravity position coordinates; The stress value time series curve and the rate of change of the centroid position coordinate over time are correlated with the monitoring time, and the change in stress-displacement relationship is calculated as a crack propagation parameter.

[0045] Stress concentration areas are determined using a preset stress threshold. An area is considered a stress concentration area when the stress values ​​measured by multiple sensors within that area exceed the preset threshold. The stress values ​​and location coordinates of the sensor nodes within the stress concentration area are recorded. For example, if a sensor is located at a specific spatial coordinate, the corresponding stress value is measured. The stress values ​​of each sensor node are sorted according to the monitoring time point, forming a stress value time series curve. This process is repeated for all sensor nodes within the stress concentration area to obtain the stress value time series curves for all nodes within that area.

[0046] The stress distribution weight of a stress concentration region is calculated based on the stress values ​​of the grid nodes. This weight reflects the relative importance of each node in the regional stress distribution; the higher the stress value, the greater the weight. The weight is calculated as the ratio of the node's stress value to the sum of the stress values ​​of all nodes within the stress concentration region. By calculating the stress distribution weight, the contribution of each node to the overall stress distribution can be quantified.

[0047] The centroid coordinates of the stress concentration region are calculated based on the grid node coordinates and stress distribution weights. The centroid coordinates are calculated by multiplying the spatial coordinate components of each node by their corresponding weights. This weighted average method provides a more accurate reflection of the spatial distribution characteristics of the stress concentration region, with higher weights given to areas of high stress.

[0048] During continuous monitoring, the centroid coordinates of the stress concentration area are calculated at each monitoring time point, forming a data sequence of centroid coordinate changes over time. By tracking the movement of the centroid coordinates, the migration of the stress concentration area over time can be observed intuitively.

[0049] The rate of change of the center of gravity coordinates over time is obtained by calculating the displacement within adjacent monitoring time intervals. The displacement is calculated as the Euclidean distance between the center of gravity coordinates at two adjacent time points, and the rate of change is the ratio of the displacement to the time interval. The rate of change of the center of gravity coordinates over time reflects the speed of migration of the stress concentration region.

[0050] The stress-displacement correlation is calculated by mapping the stress value time series curve to the rate of change of the centroid position coordinates over time, corresponding to the monitoring time, and used as a crack propagation parameter. This correlation reflects the degree of relationship between stress and displacement changes. The calculation method is the ratio of stress change to displacement change at the corresponding monitoring time point. Analyzing the changes in the stress-displacement correlation across continuous monitoring time points allows for the determination of the crack propagation rate and direction. A continuously increasing value indicates a larger stress change per unit displacement, suggesting a potential slowdown in crack propagation; conversely, a continuously decreasing value indicates a smaller stress change per unit displacement, suggesting a potential acceleration in crack propagation, warranting increased vigilance.

[0051] This invention achieves accurate acquisition of crack propagation parameters by calculating the time series curve of stress values ​​and the coordinates of the center of gravity within a stress concentration region, demonstrating significant technical advantages. This method fully utilizes stress data acquired by vibrating wire sensors, combined with spatial coordinate information, to achieve precise positioning and dynamic monitoring of stress concentration areas in tunnel structures. By introducing the concept of stress distribution weighting, the accuracy of the center of gravity location calculation is improved, more realistically reflecting the stress distribution characteristics. The change in the stress-displacement relationship, used as a crack propagation parameter, intuitively reflects the trend and speed of crack propagation, providing a reliable basis for tunnel safety early warning.

[0052] The stress-displacement relationship between the stress-time series curve and the rate of change of the centroid position coordinates over time is calculated according to the monitoring time and used as a crack propagation parameter, including: The stress value time series curve is denoised and normalized to obtain a standardized stress curve. The stress fluctuation frequency and stress fluctuation amplitude are calculated based on the standardized stress curve. The stress fluctuation frequency and stress fluctuation amplitude are combined to construct a stress displacement correspondence matrix. The displacement direction vector is calculated based on the time change rate of the center of gravity position coordinates. The displacement direction vector is mapped and matched with the stress-displacement corresponding matrix. The angle between the main direction and the main direction in the displacement direction vector and the stress-displacement corresponding matrix is ​​extracted. The displacement change intensity is calculated based on the angle between the main direction. The stress-displacement correspondence matrix is ​​corrected using the intensity of displacement change, and the stress-displacement gradient change is calculated based on the corrected stress-displacement correspondence matrix. The stress-displacement gradient change and the displacement change intensity are combined and calculated iteratively to obtain the stress-displacement relationship change, which is then used as the crack propagation parameter.

[0053] Stress data acquired by vibrating wire sensors often contains noise interference, requiring denoising through a moving average filtering method. Taking a tunnel monitoring point as an example, the original stress data fluctuates. A sliding window is set to five time points, and the average value within each window is used to replace the value at the center point, resulting in smoothed data. The smoothed data is then normalized, mapping the stress values ​​to the range of 0 to 1 to obtain a standardized stress curve.

[0054] The stress fluctuation frequency reflects the number of fluctuations in the standardized stress curve per unit time. It is calculated using the zero crossover rate method, which is the ratio of the number of times the standardized stress curve crosses its mean line to the total monitoring time. Taking the normalized data above as an example, the mean is 0.58, the curve crosses the mean line 4 times, the monitoring time is 6 hours, and the stress fluctuation frequency is 0.67 times / hour. The stress fluctuation amplitude is calculated as the standard deviation of the standardized stress curve, reflecting the severity of stress changes. The standard deviation of the normalized data above is 0.42, which is the stress fluctuation amplitude.

[0055] A stress-displacement correspondence matrix is ​​constructed by combining the stress fluctuation frequency and amplitude values. This matrix is ​​a two-dimensional data structure, where rows represent discrete intervals of stress fluctuation frequency values ​​and columns represent discrete intervals of stress fluctuation amplitude. The matrix element values ​​represent the density of data points within the corresponding interval, reflecting the probability of a specific stress fluctuation characteristic occurring. The stress fluctuation frequency range of 0-2 times / h is divided into 10 equal intervals, and the stress fluctuation amplitude range of 0-1 is also divided into 10 equal intervals, forming a 10×10 matrix. Taking the calculated stress fluctuation frequency of 0.67 times / h and stress fluctuation amplitude of 0.42 as an example, the element value in the 4th row and 5th column of the matrix is ​​increased by 1. Statistical analysis of all monitoring data yields the complete stress-displacement correspondence matrix.

[0056] The displacement direction vector is calculated based on the rate of change of the center of gravity coordinates over time. The displacement direction vector is defined as the direction of change in the center of gravity coordinates between two adjacent monitoring time points. Taking a monitoring area in a tunnel as an example, the center of gravity coordinates at two consecutive time points are (13.65, 15.31, 5.39) and (13.72, 15.35, 5.42), with a coordinate difference of (0.07, 0.04, 0.03). After normalization, the displacement direction vector is obtained as (0.82, 0.47, 0.35). This calculation is performed for each pair of adjacent time points to obtain a sequence of displacement direction vectors.

[0057] Principal component analysis (PCA) is performed on the stress-displacement matrix to obtain the principal direction, which represents the direction of most significant change in the matrix data. Eigenvalue decomposition is performed on the 10×10 stress-displacement matrix, and the eigenvector with the largest eigenvalue is the principal direction, for example (0.75, 0.62, 0.23). The angle between the displacement direction vector and the principal direction is calculated using the inverse cosine of the ratio of the inner product of the two vectors to the product of their magnitudes. Taking the displacement direction vector (0.82, 0.47, 0.35) and the principal direction (0.75, 0.62, 0.23) as an example, the calculated angle is 15.4 degrees.

[0058] The intensity of displacement change is calculated based on the included angle of the principal direction. The intensity of displacement change is defined as a measure of the consistency between the displacement direction and the principal direction, calculated by multiplying the cosine of the included angle by the magnitude of the displacement. The smaller the included angle, the greater the intensity of displacement change, reflecting a more pronounced trend of crack propagation along the principal direction. Taking the aforementioned 15.4-degree included angle as an example, the cosine value is 0.96, the magnitude of the displacement (Euclidean distance) is 0.085m, and the intensity of displacement change is 0.082m.

[0059] The stress-displacement correspondence matrix is ​​corrected using the intensity of displacement change and a Gaussian function to obtain the corrected matrix. The stress-displacement gradient change is then calculated based on this corrected matrix. The stress-displacement gradient change is defined as the rate of change of each element's value within adjacent intervals, calculated using the central difference method. For a 10×10 matrix, the difference between each element and its four adjacent elements (up, down, left, and right) is calculated, and the average value is taken as the gradient change at that point. The average of the gradient changes for all elements is then used to obtain the overall stress-displacement gradient change. In this embodiment, the calculated stress-displacement gradient change is 0.23.

[0060] The stress-displacement gradient change and displacement intensity are combined and iteratively calculated to obtain the stress-displacement relationship change. The combination is performed using linear superposition with a weighting ratio of 7:3. Taking the stress-displacement gradient change of 0.23 and displacement intensity of 0.082m as an example, the initial stress-displacement relationship change is 0.23 × 0.7 + 0.082 × 0.3 = 0.186. A threshold of 0.001 is set for the iterative calculation process; the calculation stops when the difference between two adjacent iterations is less than the threshold. After 5 iterations, a convergence value of 0.193 is obtained, which is the stress-displacement relationship change. This value is used as a crack propagation parameter to assess the safety status of the tunnel structure.

[0061] This invention achieves accurate assessment of crack propagation trends through in-depth analysis of stress time series data and by combining the changing characteristics of the center of gravity coordinates. By introducing the concepts of stress fluctuation frequency and amplitude, a stress-displacement correspondence matrix is ​​constructed, enabling comprehensive capture of stress variation characteristics. The calculation of the angle between the displacement direction vector and the principal direction effectively quantifies the directionality of crack propagation. The introduction of displacement change intensity and stress-displacement gradient change improves the sensitivity and accuracy of crack propagation parameters. It avoids the problem of over-reliance on single parameters in traditional monitoring methods, achieving multi-dimensional data fusion analysis and providing more comprehensive and reliable technical support for tunnel structural health monitoring, thus possessing significant practical value for tunnel safety early warning and maintenance.

[0062] The bearing capacity attenuation curve of the surrounding rock is calculated based on the fracture propagation parameters. The bearing capacity attenuation threshold of the surrounding rock is determined according to the stress value in the stress concentration area. When the slope of the bearing capacity attenuation curve of the surrounding rock exceeds the bearing capacity attenuation threshold of the surrounding rock, an early warning signal is output, including: Acquire temporal variation data of fracture propagation parameters, calculate the bearing capacity variation trend within the monitoring period based on the temporal variation data, and generate a bearing capacity attenuation curve of surrounding rock. Collect stress value monitoring data in the stress concentration area, calculate the change range of the stress value monitoring data, and determine the bearing capacity attenuation threshold of the surrounding rock based on the change range of the stress value; The rate of change of the surrounding rock bearing capacity at adjacent monitoring times is calculated based on the bearing capacity decay curve of the surrounding rock and used as the slope of the bearing capacity decay curve. The slope of the bearing capacity attenuation curve of the surrounding rock is compared with the bearing capacity attenuation threshold of the surrounding rock. When the slope of the bearing capacity attenuation curve of the surrounding rock exceeds the bearing capacity attenuation threshold of the surrounding rock, an early warning signal is output.

[0063] The time-series variation data of fracture propagation parameters refers to the sequence of fracture propagation parameters obtained at different monitoring times. There is a correlation between fracture propagation parameters and the bearing capacity of the surrounding rock. These parameters are converted into bearing capacity values ​​using an exponential function mapping. The basic form of this exponential function is an exponential decay function, where the initial bearing capacity value is related to the surrounding rock type, and the decay coefficient is calibrated based on field measurement data. Plotting these bearing capacity values ​​against the corresponding monitoring time forms the bearing capacity decay curve. The curve is relatively flat initially, but shows an accelerating downward trend as the monitoring time increases, reflecting the characteristic of the bearing capacity of the surrounding rock gradually decreasing during fracture development.

[0064] Stress monitoring data is collected within stress concentration areas, and the magnitude of stress value changes is calculated. The attenuation threshold of the surrounding rock bearing capacity is determined based on this magnitude. Stress monitoring data is acquired in real-time via a vibrating wire sensor network. The magnitude of stress value changes is calculated by averaging the absolute values ​​of the stress differences between measuring points at two adjacent monitoring time points. A dynamic threshold setting method is used to determine the attenuation threshold of the surrounding rock bearing capacity based on the magnitude of stress value changes. When the magnitude of stress value changes increases drastically within a short period, it indicates that the surrounding rock may be in a state of impending instability, requiring a reduction in the attenuation threshold to improve early warning sensitivity. The threshold calculation comprehensively considers the current stress change magnitude and the previous rate of change. A sliding window method is used to smooth the recent stress change magnitude data, and 1.2 times the smoothed result is taken as the attenuation threshold for the current monitoring point.

[0065] The rate of change of surrounding rock bearing capacity is defined as the amount of change in surrounding rock bearing capacity per unit time. It is calculated as the ratio of the difference in bearing capacity values ​​between two adjacent monitoring times to the time interval. These rate of change values ​​represent the slope of the surrounding rock bearing capacity decay curve at each monitoring time, reflecting the rate of decay of the bearing capacity. The larger the slope, the faster the bearing capacity decreases, and the higher the potential risk to the tunnel structure.

[0066] The slope of the rock bearing capacity decay curve is compared with the rock bearing capacity decay threshold. When the slope exceeds the threshold, an early warning signal is output. The comparison process uses a dynamic tracking method, comparing the slope value at each monitoring point with the corresponding threshold. In the example above, the slope values ​​at the first nine monitoring points are all less than the corresponding thresholds. The slope at the tenth monitoring point is 1.20 MPa / h, and the threshold is 1.26 MPa / h, still within the threshold. The slope at the eleventh monitoring point is 1.45 MPa / h, and the threshold is 1.58 MPa / h, also within the threshold. If monitoring continues, assuming the rock bearing capacity at the twelfth monitoring point is 70.1 MPa, the slope at the twelfth time period is 1.60 MPa / h, and the threshold is 1.58 MPa / h. At this point, the slope exceeds the threshold, triggering an early warning signal output. The early warning signal output includes both audible and visual alarms and remote data transmission. The audible and visual alarms use on-site sirens to emit sound and flashing lights to alert on-site personnel. The remote data transmission sends the warning information to the monitoring center and relevant management personnel via communication equipment. This information includes key details such as the warning time, location, rock bearing capacity, and slope, enabling decision-makers to understand the tunnel's condition in a timely manner and take appropriate measures.

[0067] After an early warning signal is issued, the monitoring system will automatically increase the data collection frequency from once every two hours to once every 30 minutes to obtain more intensive monitoring data and achieve precise monitoring of dangerous areas. Simultaneously, the historical data backtracking analysis function will be activated to compare the current warning situation with historical cases, assess the risk level and development trend, and provide decision support for emergency response. Early warning signals are divided into three levels: a yellow warning indicates that the slope slightly exceeds the threshold, suggesting strengthened monitoring; an orange warning indicates that the slope exceeds the threshold by more than 20%, suggesting reinforcement measures; and a red warning indicates that the slope exceeds the threshold by more than 50%, suggesting emergency evacuation and comprehensive response. Based on the warning level, relevant management personnel can take different levels of response measures to effectively prevent tunnel safety accidents.

[0068] This invention establishes a mapping relationship between the bearing capacity of surrounding rock and crack propagation through in-depth analysis of time-series data of crack propagation parameters, achieving an accurate characterization of the bearing capacity decay process. The dynamic threshold setting mechanism adaptively adjusts the early warning sensitivity based on stress changes, avoiding false alarms and missed alarms caused by fixed thresholds. The slope calculation method is simple and effective, capable of keenly capturing abrupt changes in the bearing capacity of surrounding rock. The combination of a multi-level early warning mechanism and emergency response strategy forms a complete safety assurance system. The integration of the high-precision measurement capabilities of vibrating wire sensors with intelligent data analysis technology significantly improves the accuracy and timeliness of tunnel monitoring.

[0069] like Figure 3 As shown in the figure, an embodiment of the present invention provides a tunnel monitoring system based on a vibrating wire sensor, the system comprising: The data acquisition module is used to collect vibration frequency data and spatial position coordinates of the vibrating wire sensor within the monitoring area; The stress wave field construction module is used to extract the spectral energy density distribution based on vibration frequency data, calculate the time difference and energy attenuation ratio of the spectral energy density distribution between adjacent vibrating wire sensors, and construct the stress wave propagation field of the monitoring area by combining spatial location coordinates. The stress field analysis module is used to divide the stress calculation grid within the stress wave propagation field of the monitoring area, calculate the stress value of the grid node based on the spectral energy density distribution and energy attenuation ratio, generate a stress field distribution cloud map, and extract the region with the maximum stress gradient in the stress field distribution cloud map as the stress concentration region. The crack parameter calculation module is used to calculate the stress value time series curve in the stress concentration area, extract the centroid position coordinates of the stress concentration area, and determine the crack propagation parameters based on the time change rate of the centroid position coordinates and the stress value time series curve. The early warning judgment module is used to calculate the bearing capacity attenuation curve of the surrounding rock based on the crack propagation parameters, determine the bearing capacity attenuation threshold of the surrounding rock according to the stress value in the stress concentration area, and output an early warning signal when the slope of the bearing capacity attenuation curve of the surrounding rock exceeds the bearing capacity attenuation threshold of the surrounding rock.

[0070] like Figure 4 As shown, an embodiment of the present invention also provides an electronic device, including: a memory 402, a processor 401, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in any of the aforementioned methods.

[0071] like Figure 5 As shown, an embodiment of the present invention also provides a computer-readable storage medium 501, on which a computer program is stored, and when the processor executes the computer program, it implements the steps in any of the aforementioned methods.

[0072] The specific embodiments described above are preferred embodiments of the present invention and are not intended to limit the specific scope of the present invention. The scope of the present invention includes, but is not limited to, these specific embodiments. All equivalent changes made in accordance with the shape and structure of the present invention are within the protection scope of the present invention.

Claims

1. A tunnel monitoring method based on a vibrating wire sensor, characterized in that, Includes the following steps: Collect vibration frequency data and spatial location coordinates of the vibrating wire sensor within the monitoring area; Based on vibration frequency data, the spectral energy density distribution is extracted, the time difference and energy attenuation ratio of the spectral energy density distribution between adjacent vibrating wire sensors are calculated, and the stress wave propagation field of the monitoring area is constructed by combining spatial location coordinates. Within the stress wave propagation field of the monitoring area, a stress calculation grid is divided. The stress values ​​of the grid nodes are calculated based on the spectral energy density distribution and energy attenuation ratio. A stress field distribution cloud map is generated, and the region with the maximum stress gradient in the stress field distribution cloud map is extracted as the stress concentration region. Calculate the stress value time series curve within the stress concentration region, extract the centroid position coordinates of the stress concentration region, and determine the crack propagation parameters based on the time change rate of the centroid position coordinates and the stress value time series curve. The bearing capacity attenuation curve of the surrounding rock is calculated based on the crack propagation parameters. The bearing capacity attenuation threshold of the surrounding rock is determined according to the stress value in the stress concentration area. When the slope of the bearing capacity attenuation curve of the surrounding rock exceeds the bearing capacity attenuation threshold of the surrounding rock, an early warning signal is output.

2. The method according to claim 1, characterized in that, Based on vibration frequency data, the spectral energy density distribution is extracted, and the time difference and energy attenuation ratio of the spectral energy density distribution between adjacent vibrating wire sensors are calculated. Combined with spatial location coordinates, the stress wave propagation field of the monitoring area is constructed, including: The vibration frequency data is segmented according to the sampling interval, and the segmented vibration frequency data is processed by a window function to obtain the window function processed signal. The power spectral density is obtained by performing spectral analysis on the signal processed by the window function, and the power spectral density is normalized to obtain the spectral energy density distribution. The cross-correlation coefficient between adjacent vibrating wire sensors is calculated based on the spectral energy density distribution. The time shift corresponding to the maximum value of the cross-correlation coefficient is extracted as the spectral energy density distribution time difference, and the amplitude ratio of the maximum value is extracted as the energy attenuation ratio. The wave velocity distribution is obtained by mapping the spatial location coordinates and the time difference of the spectral energy density distribution to the monitoring area. The wave velocity distribution is then corrected by combining the energy attenuation ratio to generate the stress wave propagation field in the monitoring area.

3. The method according to claim 1, characterized in that, Within the stress wave propagation field of the monitoring area, a stress calculation grid is divided. The stress values ​​at the grid nodes are calculated based on the spectral energy density distribution and energy attenuation ratio. A stress field distribution cloud map is generated, and the region with the maximum stress gradient in the stress field distribution cloud map is extracted as the stress concentration region, including: A grid coordinate system is established within the stress wave propagation field of the monitoring area. The stress calculation grid is divided according to the distribution range of the vibrating wire sensor, and the spatial coordinates of the grid nodes in the stress calculation grid are calibrated. The spectral energy density distribution is allocated based on the spatial coordinates of the grid nodes to obtain the grid node energy distribution. The attenuation coefficient of the grid node energy distribution is calculated based on the energy attenuation ratio. The stress values ​​of the grid nodes are calculated using the energy distribution and attenuation coefficient of the grid nodes, and a stress field distribution cloud map is generated based on the stress values ​​of the grid nodes. The stress value change of adjacent grid nodes in the stress field distribution cloud map is calculated, and the region where the stress value change is greater than the change of the surrounding grid nodes is determined as the region with the maximum stress gradient, which is used as the stress concentration region.

4. The method according to claim 3, characterized in that, The spectral energy density distribution is allocated based on the spatial coordinates of the grid nodes to obtain the grid node energy distribution. The attenuation coefficient of the grid node energy distribution is calculated based on the energy attenuation ratio, including: Calculate the spatial coordinate distance matrix between the grid nodes and the vibrating wire sensor, divide the spectral energy density distribution into components based on the spatial coordinate distance matrix, and determine the propagation path of the spectral energy density distribution at the spatial coordinates of the grid nodes; Obtain the angle between the propagation path and the normal direction of the tunnel wall, calculate the energy propagation loss based on the angle, correct the energy propagation loss and the spectral energy density distribution, and superimpose the corrected spectral energy density distribution at the spatial coordinates of the grid node to obtain the grid node energy distribution; Calculate the propagation distance between adjacent grid nodes, obtain the propagation attenuation factor based on the propagation distance, and use the propagation attenuation factor to correct the energy attenuation ratio to obtain the grid node energy distribution attenuation coefficient.

5. The method according to claim 1, characterized in that, Calculating the stress value time series curve within the stress concentration region, extracting the centroid coordinates of the stress concentration region, and determining the crack propagation parameters based on the time change rate of the centroid coordinates and the stress value time series curve include: Obtain the stress values ​​and coordinates of grid nodes within the stress concentration area, and construct a stress value time series curve by sorting the grid node stress values ​​according to the monitoring time. The stress distribution weight of the stress concentration region is calculated based on the stress values ​​of the grid nodes, and the centroid coordinates of the stress concentration region are calculated based on the grid node position coordinates and the stress distribution weights. Record the change data of the center of gravity position coordinates with the monitoring time, and calculate the displacement of the center of gravity position coordinates within adjacent monitoring time intervals to obtain the time change rate of the center of gravity position coordinates; The stress value time series curve and the rate of change of the centroid position coordinate over time are correlated with the monitoring time, and the change in stress-displacement relationship is calculated as a crack propagation parameter.

6. The method according to claim 5, characterized in that, The stress-displacement relationship between the stress-time series curve and the rate of change of the centroid position coordinates over time is calculated according to the monitoring time and used as a crack propagation parameter, including: The stress value time series curve is denoised and normalized to obtain a standardized stress curve. The stress fluctuation frequency and stress fluctuation amplitude are calculated based on the standardized stress curve. The stress fluctuation frequency and stress fluctuation amplitude are combined to construct a stress displacement correspondence matrix. The displacement direction vector is calculated based on the time change rate of the center of gravity position coordinates. The displacement direction vector is mapped and matched with the stress-displacement corresponding matrix. The angle between the main direction and the main direction in the displacement direction vector and the stress-displacement corresponding matrix is ​​extracted. The displacement change intensity is calculated based on the angle between the main direction. The stress-displacement correspondence matrix is ​​corrected using the intensity of displacement change, and the stress-displacement gradient change is calculated based on the corrected stress-displacement correspondence matrix. The stress-displacement gradient change and the displacement change intensity are combined and calculated iteratively to obtain the stress-displacement relationship change, which is then used as the crack propagation parameter.

7. The method according to claim 1, characterized in that, The bearing capacity attenuation curve of the surrounding rock is calculated based on the fracture propagation parameters. The bearing capacity attenuation threshold of the surrounding rock is determined according to the stress value in the stress concentration area. When the slope of the bearing capacity attenuation curve of the surrounding rock exceeds the bearing capacity attenuation threshold of the surrounding rock, an early warning signal is output, including: Acquire temporal variation data of fracture propagation parameters, calculate the bearing capacity variation trend within the monitoring period based on the temporal variation data, and generate a bearing capacity attenuation curve of surrounding rock. Collect stress value monitoring data in the stress concentration area, calculate the change range of the stress value monitoring data, and determine the bearing capacity attenuation threshold of the surrounding rock based on the change range of the stress value. The rate of change of the surrounding rock bearing capacity at adjacent monitoring times is calculated based on the bearing capacity decay curve of the surrounding rock and used as the slope of the bearing capacity decay curve. The slope of the bearing capacity attenuation curve of the surrounding rock is compared with the bearing capacity attenuation threshold of the surrounding rock. When the slope of the bearing capacity attenuation curve of the surrounding rock exceeds the bearing capacity attenuation threshold of the surrounding rock, an early warning signal is output.

8. A tunnel monitoring system based on a vibrating wire sensor, used to implement the method described in any one of claims 1-7, characterized in that, The system includes: The data acquisition module is used to collect vibration frequency data and spatial position coordinates of the vibrating wire sensor within the monitoring area; The stress wave field construction module is used to extract the spectral energy density distribution based on vibration frequency data, calculate the time difference and energy attenuation ratio of the spectral energy density distribution between adjacent vibrating wire sensors, and construct the stress wave propagation field of the monitoring area by combining spatial location coordinates. The stress field analysis module is used to divide the stress calculation grid within the stress wave propagation field of the monitoring area, calculate the stress value of the grid node based on the spectral energy density distribution and energy attenuation ratio, generate a stress field distribution cloud map, and extract the region with the maximum stress gradient in the stress field distribution cloud map as the stress concentration region. The crack parameter calculation module is used to calculate the stress value time series curve in the stress concentration area, extract the centroid position coordinates of the stress concentration area, and determine the crack propagation parameters based on the time change rate of the centroid position coordinates and the stress value time series curve. The early warning judgment module is used to calculate the bearing capacity attenuation curve of the surrounding rock based on the crack propagation parameters, determine the bearing capacity attenuation threshold of the surrounding rock according to the stress value in the stress concentration area, and output an early warning signal when the slope of the bearing capacity attenuation curve of the surrounding rock exceeds the bearing capacity attenuation threshold of the surrounding rock.

9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions that, when executed by a processor, implement the steps of the method as described in any one of claims 1 to 7.

Images