Monitoring and early warning method and system for underground engineering cracks

Abstract

The invention relates to the technical field of intelligent monitoring, in particular to a monitoring and early warning method and system for underground engineering cracks, and the method comprises the steps: determining an area with cracks on the surface of surrounding rock, obtaining the static characteristics of the surrounding rock cracks, and obtaining the width, displacement and surrounding rock stress distribution data of the cracks; establishing a constitutive model of the fractured rock mass, and establishing a three-dimensional geometric model of the surrounding rock based on engineering geometric parameters; taking crack width and displacement data obtained through detection as input characteristics, performing inversion on the three-dimensional geometric model based on a deep learning model, and outputting crack width and displacement rate in future time; and establishing a graded early warning mechanism, determining standard parameters of each stage of early warning mechanism, comparing the output crack width and displacement rate prediction value with the standard parameters for judgment, and performing monitoring and early warning. The crack width and the corresponding displacement rate in the future time can be obtained, monitoring and early warning are carried out according to the development trend of the predicted crack, and potential safety hazards are found and processed in time.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent monitoring, and specifically provides a monitoring and early warning method and system for cracks in underground engineering. Background Art

[0002] During the construction of underground engineering, due to the complexity of geological structures, excavation operation disturbances, and the influence of external loads, crack problems are likely to occur. The expansion of these cracks may lead to the instability of surrounding rocks, water seepage, and even collapse, which undoubtedly pose a threat to the safety of the entire underground engineering. Therefore, monitoring and early warning of cracks can minimize safety losses.

[0003] However, the existing monitoring methods for cracks mainly rely on manual inspections or sensor detections, but there are still some problems:

[0004] The existing crack monitoring means are inefficient: Traditional crack monitoring relies on manual inspections or single sensors, and there are problems such as limited coverage, lagging data updates, and poor real-time performance.

[0005] Insufficient real-time performance and accuracy: Existing methods are difficult to capture the minute changes of cracks in real time, and have weak prediction capabilities for the crack expansion trend, resulting in delayed early warnings.

[0006] Difficulty in multi-source data fusion: The underground engineering environment is complex, and the development of cracks is affected by multiple factors such as geology, stress, and construction. Existing technologies lack the ability of collaborative analysis and dynamic modeling of multi-source data. Summary of the Invention

[0007] To solve the above problems, on the one hand, the present invention provides a monitoring and early warning method for cracks in underground engineering, including:

[0008] S1. Determine the area where cracks exist on the surface of the surrounding rock, scan the internal structure of the surrounding rock cracks through a geological radar sensor to obtain the static characteristics of the surrounding rock cracks; and set the sensor device at the position of the surrounding rock cracks to detect and obtain data on crack width, displacement, and surrounding rock stress distribution;

[0009] S2. Establish a constitutive model of the cracked rock mass, and map the static characteristics of the cracks into the constitutive model; based on the engineering geometric parameters, establish a three-dimensional geometric model of the surrounding rock, import the constitutive model of the cracked rock mass and the surrounding rock stress distribution data, and simulate the stress distribution around the surrounding rock cracks;

[0010] S3. Use the detected crack width and displacement data as input features, and perform inversion on the three-dimensional geometric model based on the deep learning model to output the crack width and the corresponding displacement rate at a future time;

[0011] S4. Establish a hierarchical early warning mechanism and determine the standard parameters of each level of the early warning mechanism. Compare and judge the predicted values of the crack width and displacement rate obtained by the output with the standard parameters to conduct monitoring and early warning.

[0012] The constitutive model of the cracked rock mass adopts an elastoplastic-damage coupling model, specifically including:

[0013] In the elastic stage, the generalized Hooke's law is used to describe the initial deformation of the surrounding rock.

[0014] In the plastic stage, the Mohr-Coulomb criterion is introduced to judge shear failure.

[0015] For damage evolution, a damage variable D is defined to characterize the crack density and propagation rate:

[0016]

[0017] Among them, σ eff is the equivalent stress; σ c is the critical stress; α is the rock mass material property parameter, and n is a constant.

[0018] The three-dimensional geometric model of the surrounding rock established based on the engineering geometric parameters in S2 is specifically as follows:

[0019] Obtain the geometric point cloud data of the underground project, including the excavation surface, the support structure, and the surface of the surrounding rock; collect the geological borehole data, extract the depth, dip angle, and lithology parameters of the rock layer interface; import all the geometric parameters of the designed underground project into the BIM platform, and establish a three-dimensional geometric model of the surrounding rock through the BIM platform.

[0020] The operation of using the crack width and displacement data obtained by detection as input features after preprocessing in S3 is specifically as follows:

[0021] Obtain the relative displacement on both sides of the crack based on the sampling frequency of the sensor, and obtain the dynamic value of the crack width based on the relative displacement and the static feature of the crack width;

[0022] Perform a first-order difference on the dynamic values of the crack width in a continuous time period to obtain the instantaneous rate;

[0023] Obtain the detection data at several sampling frequencies, and determine the crack width data set and the displacement rate data set as input features.

[0024] To ensure the accuracy of the predicted value of the crack width, S3 also includes verifying the stress intensity factor of the predicted value of the crack width. If the stress intensity at the crack tip is greater than or equal to the fracture toughness of the rock mass, the predicted value of the crack width is corrected.

[0025] The verification of the stress intensity factor for the predicted value of the crack width is specifically as follows:

[0026] Input the predicted crack width value into the finite element analysis model to calculate the stress intensity factor K at the crack tip I , and its formula is:

[0027]

[0028] where σ represents the far - field stress perpendicular to the crack surface; a represents the half - length of the crack; Y is the geometric correction factor; W 预测 represents the predicted crack width value; x represents the distance from a point on the crack surface to the center; E is the elastic modulus; v represents the Poisson's ratio.

[0029] The correction of the predicted crack width value is carried out according to the formula:

[0030]

[0031] In the formula, α is the rock mass material property parameter; K I is the stress intensity factor at the crack tip; K IC is the rock mass fracture toughness.

[0032] In the specific implementation, the hierarchical early - warning mechanism is specifically divided into an observation level, a warning level, and an emergency level.

[0033] In order to achieve the accuracy of crack monitoring and early warning, the predicted crack width and displacement rate obtained in S4 are compared with the standard parameters for monitoring and early warning, specifically as follows:

[0034] If the predicted crack width value ≥ 2mm or the predicted displacement rate value ≥ 0.1mm / h, it reaches the observation level and triggers a yellow early - warning;

[0035] If the predicted crack width value ≥ 4mm and the predicted displacement rate values in each future time period increase successively, it reaches the warning level and triggers an orange early - warning;

[0036] If the predicted crack width value ≥ 6mm or the predicted displacement rate value ≥ 1mm / 10min, it reaches the warning level and triggers a red early - warning.

[0037] On the other hand, the present invention provides a monitoring and early - warning system for cracks in underground engineering, including:

[0038] A data acquisition module, used to determine the area where cracks exist on the surrounding rock surface, scan the internal structure of the surrounding rock cracks through a ground - penetrating radar sensor to obtain the static characteristics of the surrounding rock cracks; and set sensor devices at the positions of the surrounding rock cracks to detect and obtain data on crack width, displacement, and surrounding rock stress distribution;

[0039] A model module for establishing a constitutive model of fractured rock mass, mapping the static characteristics of fractures into the constitutive model; establishing a three-dimensional geometric model of the surrounding rock based on engineering geometric parameters, importing the constitutive model of the fractured rock mass and the surrounding rock stress distribution data, and simulating the stress distribution around the fractures in the surrounding rock.

[0040] A prediction module for using the crack width and displacement data obtained by detection as input features, inversely analyzing the three-dimensional geometric model based on a deep learning model, and outputting the crack width and corresponding displacement rate at a future time.

[0041] A monitoring and early warning module for establishing a hierarchical early warning mechanism and determining the standard parameters of each level of the early warning mechanism, comparing and judging the predicted values of the crack width and displacement rate obtained by output with the standard parameters, and conducting monitoring and early warning.

[0042] Beneficial effects: The present invention is a method and system for monitoring and early warning of cracks in underground engineering. By deploying relevant sensor devices in the underground engineering structure, a dense sensor network is formed, which can accurately obtain the static characteristics of the fractures in the surrounding rock, as well as the crack width, displacement, and surrounding rock stress distribution data, providing a reliable basis for subsequent analysis and prediction.

[0043] Using a deep learning model to inversely predict the geometric model of the fractures in the surrounding rock can obtain the crack width and corresponding displacement rate at a future time, improving the accuracy and timeliness of crack monitoring and early warning; the present invention also checks and corrects the stress intensity factor of the predicted crack width, further ensuring the accuracy of the prediction results; at the same time, a hierarchical early warning mechanism is established and the standard parameters of each level of the early warning mechanism are determined, monitoring and early warning are carried out according to the development trend of the predicted cracks, and an early warning signal is sent when the crack reaches the early warning threshold, which can timely discover and handle potential safety hazards, avoid accidents, and ensure the safe operation of underground engineering. Description of the Drawings

[0044] By reading the detailed description of the preferred embodiments below, the solutions and advantages of the present application will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention.

[0045] Figure 1 It is a flowchart of a method for monitoring and early warning of cracks in underground engineering. Detailed Embodiments

[0046] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the drawings.

[0047] Embodiment

[0048] See Figure 1, this embodiment provides a monitoring and early warning method for cracks in underground engineering. The operation steps of the method are as follows:

[0049] S1. Determine the area where cracks exist on the surrounding rock surface. Scan the internal structure of the surrounding rock cracks through a ground penetrating radar sensor to obtain the static characteristics of the surrounding rock cracks; and set the sensor device at the position of the surrounding rock cracks to detect and obtain data on crack width, displacement, and surrounding rock stress distribution.

[0050] First, for the area where cracks exist on the surface of the surrounding rock of the underground engineering, scan the internal structure of the surrounding rock cracks through a ground penetrating radar sensor to obtain the static data of the surrounding rock cracks. The static data includes crack shape, length, width, and depth.

[0051] Set relevant sensor devices in the key areas (such as the vault and side walls) of the surrounding rock cracks. The sensor devices include fiber Bragg grating sensors, acoustic emission sensors, and temperature and humidity sensors. The fiber Bragg grating sensors are used to detect data on crack width, displacement, and surrounding rock stress distribution; the acoustic emission sensors are used to capture acoustic wave signals during the crack propagation process and identify the high-frequency energy release time; the temperature and humidity sensors are used to collect on-site temperature and humidity values.

[0052] S2. Establish a constitutive model of the cracked rock mass, and map the static characteristics of the cracks into the constitutive model; based on the engineering geometric parameters, establish a three-dimensional geometric model of the surrounding rock, import the constitutive model of the cracked rock mass and the surrounding rock stress distribution data, and simulate the stress distribution around the surrounding rock cracks.

[0053] The constitutive model of the cracked rock mass adopts an elastoplastic-damage coupling model, which specifically includes:

[0054] Elastic stage: The generalized Hooke's law is used to describe the initial deformation of the surrounding rock.

[0055] Plastic stage: The Mohr-Coulomb criterion is introduced to judge shear failure.

[0056] Damage evolution: Define the damage variable D to characterize the crack density and propagation rate:

[0057]

[0058] where, σ eff is the equivalent stress; σ c is the critical stress; α is the rock mass material characteristic parameter, and n is a constant.

[0059] At the same time, perform a multi-field coupling expansion on the constitutive model of the cracked rock mass, considering the seepage-stress coupling effect, and its formula is:

[0060]

[0061] Among them, k is the permeability; p is the pore pressure, and φ is the porosity.

[0062] Meanwhile, obtain the geometric point cloud data of the underground project, including the excavation face, the support structure, and the surrounding rock surface; collect the geological borehole data, extract the depth, dip angle, and lithology parameters of the rock layer interface; import all the geometric parameters of the designed underground project into the BIM platform (Building Information Modeling), and establish a three-dimensional geometric model of the surrounding rock through the BIM platform.

[0063] Import the constitutive model of the fractured rock mass into the established three-dimensional geometric model of the surrounding rock, and simulate the stress distribution around the fractures of the surrounding rock by importing the stress distribution data of the surrounding rock detected by fiber Bragg grating sensors. Predict the development trend of the fracture by simulating the stress distribution changes caused by the underground project.

[0064] S3. Use the detected crack width and displacement data as input features, and perform inversion on the three-dimensional geometric model based on the deep learning model to output the crack width and the corresponding displacement rate at future times;

[0065] The operation of preprocessing the detected crack width and displacement data and using them as input features is specifically as follows:

[0066] Obtain the relative displacement on both sides of the crack based on the sampling frequency of the fiber Bragg grating sensor, and obtain the dynamic value of the crack width based on the relative displacement and the crack width of the static feature;

[0067] Perform a first-order difference on the dynamic values of the crack width in a continuous time period to obtain the instantaneous rate as:

[0068]

[0069] Obtain the detection data of several sampling periods, and determine the crack width data set (W t-n , …, W t ) and the displacement rate data set (V t-n , …, V t ) as input features.

[0070] Construct a deep learning model for predicting the development trend of the crack. The deep learning model uses an LSTM neural network model, which includes an input layer, a hidden layer, and an output layer; the input layer is used to input the input features, including the crack width data and the displacement rate data; the hidden layer consists of 2 layers of LSTM neural networks and is used to capture the dependence of the crack width and displacement over time; the output layer is used to output the predicted values of the crack width and displacement rate at future times.

[0071] By inputting input features into an LSTM neural network model, the LSTM neural network model simulates a three-dimensional geometric model containing surrounding rock stress distribution data and outputs predicted values of crack widths for each time period within a preset future time period.

[0072] When predicted values of crack widths for future time are obtained based on the output of a deep learning model, it further includes performing a stress intensity factor verification on the predicted values of crack widths. If the stress intensity at the crack tip is greater than or equal to the rock mass fracture toughness, the predicted values of crack widths are corrected.

[0073] The stress intensity factor verification of the predicted values of crack widths is specifically as follows:

[0074] Input the predicted values of crack widths into a finite element analysis model (ABAQUS model) to calculate the stress intensity factor K at the crack tip I , K I is a key parameter characterizing the stress field intensity at the crack tip, and its calculation formula is:

[0075]

[0076] where, σ represents the far-field stress perpendicular to the crack surface; a represents the half-length of the crack; Y is the geometric correction factor.

[0077] The half-length a of the crack is obtained by conversion through geometric relations from the predicted values of crack widths, and its conversion formula is:

[0078]

[0079] In the formula, W 预测 represents the predicted values of crack widths; x represents the distance from a point on the crack surface to the center; E is the elastic modulus; v represents the Poisson's ratio.

[0080] If the stress intensity factor K at the crack tip I exceeds the rock mass fracture toughness K IC , then the predicted values of crack widths are corrected, and its correction formula is:

[0081]

[0082] In the formula, α is a rock mass material property parameter.

[0083] After the predicted values of crack widths are corrected, the displacement rate is obtained according to the corrected values of crack widths.

[0084] S4. Establish a hierarchical early warning mechanism and determine the standard parameters of each level of early warning mechanism. Compare and judge the predicted values of crack widths and displacement rates obtained with the standard parameters for monitoring and early warning.

[0085] The hierarchical early warning mechanism is specifically divided into the observation level, the warning level, and the emergency level.

[0086] The predicted values of the crack width and displacement rate obtained by output are compared with the standard parameters for monitoring and early warning. Specifically:

[0087] If the predicted value of the crack width ≥ 2 mm or the predicted value of the displacement rate ≥ 0.1 mm / h, it reaches the observation level, triggers a yellow early warning, and prompts manual on-site review.

[0088] If the predicted value of the crack width ≥ 4 mm and the predicted values of the displacement rate in each future time period increase successively, it reaches the warning level, triggers an orange early warning, and prompts manual reinforcement of the surrounding rock of the underground project.

[0089] If the predicted value of the crack width ≥ 6 mm or the predicted value of the displacement rate ≥ 1 mm / 10 min, it reaches the warning level, triggers a red early warning, and conducts an emergency early warning.

[0090] In addition, a monitoring and early warning system for cracks in underground projects is also provided, including:

[0091] A data acquisition module, which is used to determine the area where cracks exist on the surface of the surrounding rock, scan the internal structure of the surrounding rock cracks through a ground penetrating radar sensor to obtain the static characteristics of the surrounding rock cracks; and set the sensor device at the position of the surrounding rock cracks to detect and obtain data on crack width, displacement, and surrounding rock stress distribution.

[0092] A model module, which is used to establish a constitutive model of the cracked rock mass and map the static characteristics of the cracks into the constitutive model; establish a three-dimensional geometric model of the surrounding rock based on the engineering geometric parameters, import the constitutive model of the cracked rock mass and the surrounding rock stress distribution data, and simulate the stress distribution around the surrounding rock cracks.

[0093] A prediction module, which is used to take the detected crack width and displacement data as input features, perform inversion on the three-dimensional geometric model based on a deep learning model, and output the crack width and corresponding displacement rate at future times.

[0094] A monitoring and early warning module, which is used to establish a hierarchical early warning mechanism and determine the standard parameters of each level of early warning mechanism, compare and judge the predicted values of the crack width and displacement rate obtained by output with the standard parameters, and conduct monitoring and early warning.

Claims

1. A monitoring and early warning method for cracks in underground engineering, characterized in that, Including: S1. Determine the area where cracks exist on the surrounding rock surface, scan the internal structure of the surrounding rock cracks through a ground penetrating radar sensor, and obtain the static characteristics of the surrounding rock cracks; And set the sensor device at the position of the surrounding rock cracks to detect and obtain data on crack width, displacement, and surrounding rock stress distribution; S2. Establish a constitutive model of the cracked rock mass, and map the static characteristics of the cracks into the constitutive model; establish a three-dimensional geometric model of the surrounding rock based on engineering geometric parameters, import the constitutive model of the cracked rock mass and the surrounding rock stress distribution data, and simulate the stress distribution around the surrounding rock cracks; S3. Use the detected crack width and displacement data as input features, and perform inversion on the three-dimensional geometric model based on a deep learning model to output the crack width and corresponding displacement rate at future times; S4. Establish a hierarchical warning mechanism and determine the standard parameters of each level of warning mechanism, compare and judge the predicted values of the crack width and displacement rate obtained, and conduct monitoring and warning.

2. The monitoring and early warning method for cracks in underground engineering according to claim 1, wherein, The constitutive model of the cracked rock mass adopts an elastoplastic-damage coupling model, which specifically includes: In the elastic stage, the generalized Hooke's law is used to describe the initial deformation of the surrounding rock; In the plastic stage, the Mohr-Coulomb criterion is introduced to judge shear failure; For damage evolution, a damage variable D is defined to characterize the crack density and propagation rate: Among them, σ eff is the equivalent stress; σ c is the critical stress; α is the characteristic parameter of the rock mass material, and n is a constant.

3. A monitoring and early warning method for cracks in underground engineering according to claim 1, characterized in that In S2, the establishment of the three-dimensional geometric model of the surrounding rock based on engineering geometric parameters is specifically as follows: Obtain the geometric point cloud data of the underground project, including the excavation surface, support structure, and surrounding rock surface; collect geological borehole data, extract the depth, dip angle, and lithological parameters of the rock layer interface; import all the geometric parameters of the designed underground project into the BIM platform, and establish a three-dimensional geometric model of the surrounding rock through the BIM platform.

4. A monitoring and early warning method for cracks in underground engineering according to claim 1, characterized in that, In S3, the crack width and displacement data obtained by detection are preprocessed and used as input features. The specific operation is as follows: Obtain the relative displacement on both sides of the crack based on the sampling frequency of the sensor, and obtain the dynamic value of the crack width based on the relative displacement and the crack width of the static characteristics; Perform a first-order difference on the dynamic values of the crack width in a continuous time period to obtain the instantaneous rate; Obtain the detection data at several sampling frequencies, and determine the crack width data set and displacement rate data set as input features.

5. A monitoring and early warning method for cracks in underground engineering according to claim 1, characterized in that, S3 also includes verifying the stress intensity factor of the predicted crack width. If the stress intensity at the crack tip is greater than or equal to the fracture toughness of the rock mass, the predicted crack width value is corrected.

6. The monitoring and early warning method for cracks in underground engineering according to claim 5, characterized in that, The verification of the stress intensity factor for the predicted crack width is specifically as follows: Input the predicted crack width value into the finite element analysis model to calculate the stress intensity factor \(K\) at the crack tip I , and its formula is as follows: Among them, σ represents the far-field stress perpendicular to the fracture plane; a represents the half-length of the fracture; Y is the geometric correction factor; W 预测 represents the predicted value of the fracture width; x represents the distance from a certain point on the fracture plane to the center; E is the elastic modulus; v represents the Poisson's ratio.

7. A method for monitoring and early warning of cracks in underground engineering according to claim 5, characterized in that, The correction formula for the predicted crack width value is: where α is the characteristic parameter of the rock mass material; K I is the stress intensity factor at the crack tip; K IC is the fracture toughness of the rock mass.

8. A monitoring and early warning method for cracks in underground engineering according to claim 1, characterized in that, The hierarchical warning mechanism is specifically divided into an observation level, a warning level, and an emergency level.

9. A monitoring and early warning method for cracks in underground engineering according to claim 8, characterized in that, In S4, the comparison and judgment of the predicted values of the crack width and displacement rate obtained with the standard parameters for monitoring and warning are specifically as follows: If the predicted crack width value ≥ 2 mm or the predicted displacement rate value ≥ 0.1 mm / h, it reaches the observation level and triggers a yellow warning; If the predicted crack width value ≥ 4 mm and the predicted displacement rate values in each future time period increase successively, it reaches the warning level and triggers an orange warning; If the predicted crack width value ≥ 6 mm or the predicted displacement rate value ≥ 1 mm / 10 min, it reaches the warning level and triggers a red warning.

10. A monitoring and early warning system for cracks in underground engineering, characterized in that, Including: A data acquisition module, which is used to determine the area where cracks exist on the surface of surrounding rock, scan the internal structure of the surrounding rock cracks through a ground penetrating radar sensor, and obtain the static characteristics of the surrounding rock cracks; And set the sensor device at the position of the surrounding rock crack to detect and obtain the crack width, displacement and surrounding rock stress distribution data; A model module, which is used to establish a constitutive model of the cracked rock mass, map the static characteristics of the cracks into the constitutive model; establish a three-dimensional geometric model of the surrounding rock based on the engineering geometric parameters, import the constitutive model of the cracked rock mass and the surrounding rock stress distribution data, and simulate the stress distribution around the surrounding rock cracks; A prediction module, which is used to use the detected crack width and displacement data as input features, invert the three-dimensional geometric model based on a deep learning model, and output the crack width and corresponding displacement rate at a future time; A monitoring and early warning module, which is used to establish a hierarchical early warning mechanism and determine the standard parameters of each level of early warning mechanism, compare and judge the predicted values of the crack width and displacement rate obtained by output with the standard parameters, and conduct monitoring and early warning.

Images