Intelligent monitoring and early warning system and method for construction of red-bed soft rock large-section water delivery tunnel

Abstract

The invention discloses an intelligent monitoring and early warning system and method for construction of a red layer soft rock large-section water delivery tunnel, the system comprises an automatic integrated monitoring device and an intelligent management terminal, and the automatic integrated monitoring device comprises a power module, a three-dimensional laser scanning module, a depth camera module and a data storage and transmission module; the intelligent management terminal comprises a data processing module, an early warning module and a prediction module. The defect of construction monitoring of the red-bed soft rock large-section water delivery tunnel at the present stage is overcome, the accuracy of tunnel deformation monitoring and the reasonability of tunnel deformation prediction are greatly improved, and engineering construction of the red-bed soft rock large-section water delivery tunnel is better guided.

Description

Technical Field

[0001] The present invention belongs to the field of tunnel engineering, and relates to a water conveyance tunnel construction monitoring system and method, and specifically to a system and method for analyzing and predicting deformation of a large-section water conveyance tunnel in red layer soft rock. Background Art

[0002] In water conservancy projects, water transfer tunnels are a common hydraulic structure. In soft rock formations, due to the low strength of the rock and poor self-stabilization ability, collapse and deformation are prone to occur during the excavation process, which brings great difficulties and safety risks to the construction. Therefore, monitoring and measuring the stability and deformation of the surrounding rock is crucial in the construction of large-section water transfer tunnels in soft rocks. On-site measurements of surrounding rock can be used to monitor structural response, evaluate structural safety in real time, maintain the natural symmetry of the structure, determine the timing of secondary lining, prove the rationality of the selected support method, and guide the design and construction of large-section water transfer tunnels in soft rocks.

[0003] During the construction of large-section soft rock water diversion tunnels, traditional monitoring methods mainly rely on manual on-site measurement and observation, which is not only inefficient, but also difficult to ensure the accuracy and timeliness of monitoring data under complex geological conditions. Therefore, it has become an urgent need to develop a real-time, automatic, accurate, and environmentally friendly monitoring system and monitoring method to monitor the construction of the tunnel and guide the construction of the tunnel. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent monitoring and early warning system and method for the construction of a large-section water conveyance tunnel in a red layer soft rock, which is used to monitor the deformation of the tunnel during the construction of the large-section water conveyance tunnel in a red layer soft rock. The present invention overcomes the shortcomings of the current monitoring of the construction of a large-section water conveyance tunnel in a red layer soft rock, greatly improves the accuracy of tunnel deformation monitoring and the rationality of tunnel deformation prediction, and better guides the engineering construction of a large-section water conveyance tunnel in a red layer soft rock.

[0005] The objective of the present invention is achieved through the following technical solutions:

[0006] An intelligent monitoring and early warning system for the construction of a large-section water conveyance tunnel in red soft rock, comprising an automated integrated monitoring device and an intelligent management terminal, wherein:

[0007] The automated integrated monitoring device includes a power module, a three-dimensional laser scanning module, a depth camera module, and a data storage and transmission module;

[0008] The power module is used to carry and transport the three-dimensional laser scanning module, the depth camera module, and the data storage and transmission module, change the position in the water transfer tunnel, and determine the position coordinate information of the collection point;

[0009] The three-dimensional laser scanning module is used to obtain point cloud data in the tunnel;

[0010] The depth camera module is used to obtain depth information and image information;

[0011] The data storage and transmission module is connected to the 3D laser scanning module and the depth camera module, and is used to store the point cloud data collected by the 3D laser scanning module, the depth information and image information obtained by the depth camera module, and the position coordinate information of the collection point, and transmit it to the intelligent management terminal via wireless.

[0012] The intelligent management terminal includes a data processing module, an early warning module, and a prediction module;

[0013] The data processing module is connected to the data storage and transmission module and is used to process the data acquired by the three-dimensional laser scanning module and the depth camera module;

[0014] The early warning module is used to compare the result processed by the data processing module with the set value, and to issue an alarm if the threshold is exceeded;

[0015] The prediction module is used to learn the data obtained by the three-dimensional laser scanning module and the depth camera module to achieve the purpose of predicting subsequent monitoring data.

[0016] A method for intelligent monitoring and early warning of large-section water conveyance tunnel construction in red soft rock using the above detection system, using an automated integrated monitoring device to obtain information in the tunnel, and performing intelligent analysis and early warning of the tunnel situation through an intelligent management terminal, including the following steps:

[0017] Step S1: Install the 3D laser scanning module, the depth camera module, and the data storage and transmission module on the platform of the power module for self-inspection and calibration;

[0018] Step S2: placing the automated integrated monitoring device at the monitoring starting point, and issuing monitoring instructions to the automated integrated monitoring device by the intelligent management terminal to start acquiring point cloud data and image data;

[0019] Step S3: The intelligent management terminal receives the monitoring data, analyzes and processes it, and issues an early warning in a timely manner according to the processing results;

[0020] Step S4: The intelligent management terminal learns according to the monitoring data, processing results, and early warning conditions, and ultimately achieves the purpose of prediction.

[0021] Compared with the prior art, the present invention has the following advantages:

[0022] 1. The present invention can monitor the construction of water conveyance tunnels in a timely, automatic, accurate and environmentally friendly manner.

[0023] 2. The present invention utilizes 3D laser scanning technology in combination with a depth camera, and utilizes the high-precision pixel-level information unique to the image to compensate for the lack of precision in point cloud information, thereby improving fitting accuracy and the accuracy of calculating deformation.

[0024] 3. The present invention can make timely responses to large deformations of the tunnel, thereby improving the safety of tunnel construction.

[0025] 4. The present invention can learn from existing data and use the learning results to predict tunnel deformation based on subsequent monitoring data, so as to more accurately estimate the construction progress and better guide the construction of tunnel projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 Schematic diagram of the power module equipped with a monitoring module. DETAILED DESCRIPTION

[0027] The technical solution of the present invention is further described below in conjunction with the accompanying drawings, but is not limited thereto. Any modification or equivalent replacement of the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention should be included in the protection scope of the present invention.

[0028] The present invention provides an intelligent monitoring and early warning system for the construction of a large-section water conveyance tunnel in red soft rock. Figure 1 The system includes an automated integrated monitoring device and an intelligent management terminal, wherein:

[0029] The automated integrated monitoring device comprises a power module, a three-dimensional laser scanning module 1, a depth camera module 2, and a data storage and transmission module 4;

[0030] The power module is used to carry and transport the three-dimensional laser scanning module 1, the depth camera module 2, and the data storage and transmission module 4, change the position in the water transfer tunnel, and determine the position coordinate information of the collection point;

[0031] The three-dimensional laser scanning module 1 is used to obtain point cloud data in the tunnel;

[0032] The depth camera module 2 is used to obtain depth information and image information;

[0033] The data storage and transmission module 4 is connected to the 3D laser scanning module 1 and the depth camera module 2, and is used to store the point cloud data collected by the 3D laser scanning module 1, the depth information and image information obtained by the depth camera module, and the position coordinate information of the collection point, and transmit it to the intelligent management terminal via wireless;

[0034] The intelligent management terminal includes a data processing module, an early warning module, and a prediction module;

[0035] The data processing module is connected to the data storage and transmission module 4 and is used to process the data acquired by the three-dimensional laser scanning module 1 and the depth camera module 2;

[0036] The early warning module is used to compare the result processed by the data processing module with the set value, and to issue an alarm if the threshold is exceeded;

[0037] The prediction module is used to learn the data obtained by the three-dimensional laser scanning module 1 and the depth camera module 2 to achieve the purpose of predicting subsequent monitoring data.

[0038] The present invention also provides a method for intelligent monitoring and early warning of large-section water diversion tunnel construction in red soft rock using the above-mentioned detection system. The method first assembles the automated integrated monitoring device and transports it to the site. Next, instructions are sent to it through the intelligent management terminal to start acquiring information inside the tunnel. Afterwards, cubic B-spline interpolation is used to fit the tunnel point cloud data obtained by the three-dimensional laser scanner, and the data obtained by the depth camera is used to correct the fitting function and establish a real-life picture of the tunnel construction. Subsequently, the most normal vector is introduced to calculate the deformation of the tunnel, and an early warning is issued through the intelligent management terminal. Finally, a neural network is trained based on the previous data so that it can predict tunnel deformation based on the monitoring data. Specifically, the following steps are included:

[0039] Step S1: Install the 3D laser scanning module 1, the depth camera module 2, and the data storage and transmission module 4 on the platform of the power module for self-checking and calibration. The specific steps are as follows:

[0040] Step S11: Install the three-dimensional laser scanning module 1, the depth camera module 2, and the data storage and transmission module 4 on the platform of the power module, and check the firmness of their installation;

[0041] Step S12: Check the environment perception radar 5 of the power module itself to ensure that it can sense accurately and automatically avoid obstacles and stop;

[0042] Step S13: calibrate and level the three-dimensional laser scanning module 1, the depth camera module 2 and its supporting rotation mechanism, angle detection mechanism, etc.

[0043] Step S2: Place the automated integrated monitoring device at the monitoring starting point, and the intelligent management terminal issues monitoring instructions to the automated integrated monitoring device to start acquiring point cloud data and image data. The specific steps are as follows:

[0044] Step S21: After the automated integrated monitoring device is inspected, it is transported to the tunnel monitoring starting point and placed according to the location;

[0045] Step S22: Check and calibrate the 3D laser scanning module 1 and the depth camera module 2 again, and check whether the data storage and transmission module 4 can work normally and whether the data transmission is disturbed;

[0046] Step S23: The intelligent management terminal issues a monitoring instruction to the automated integrated monitoring device, and sets key information such as monitoring content, scanning speed, and stop conditions;

[0047] Step S24: the three-dimensional laser scanning module 1 and the depth camera module 2 mounted on the power module continuously acquire the point cloud data and the corresponding depth information and image information inside the tunnel as the vehicle moves forward, and the repeatedly acquired part can be used as data optimization;

[0048] Step S25: The data storage and transmission module 4 mounted on the power module synchronously transmits the data collected by the three-dimensional laser scanning module and the depth camera module, as well as the camera angle information and the collection point coordinate information to the intelligent management terminal, and backs it up to its own storage module.

[0049] Step S3: The intelligent management terminal receives the monitoring data, analyzes and processes it, and issues an early warning in a timely manner based on the processing results. The specific steps are as follows:

[0050] Step S31: the data processing module backs up the received data again;

[0051] Step S32: The data processing module performs B-spline interpolation fitting on the received point cloud data. The specific method is as follows:

[0052] The cubic B-spline interpolation method is a process of mathematically solving the three moment equations through a smooth curve with a series of shape value points to obtain a curve function group. In the two-dimensional plane curve interpolation, the nodes x0, x1, ..., x n The cubic spline function on is S(x)∈C2[a,b]; in each small interval [x j ,x j+1 ] is a cubic polynomial, where a=x0<x1<…<x n = b defines the recipient node. If at node x j Given function value Y j =f(X j )(j=0,1,…,n), establish the cubic spline function as S(x j )=y j (j=0,1,…,n), then S(x) is called cubic spline interpolation function.

[0053] For m+n+1 scattered points in space, Q h (h=0,1,2,…,m+n) represents the set of these scattered points. Therefore, the nth spline curve can be defined as follows:

[0054]

[0055] Among them k,n (t) is the basis function of the n-order spline function, and t represents the node parameter, that is, t∈[0,1].

[0056] In spline interpolation, when n = 3, k = 0, 1, 2, 3, the polygon can be represented by 4 points, namely Q0, Q1, Q2, Q3. The cubic spline is a cubic polynomial defined as follows:

[0057]

[0058]

[0059] Therefore, the expression for the cubic spline is:

[0060]

[0061] Where i = 0, 1, 2,…, m.

[0062] According to the principle of cubic spline interpolation in two-dimensional space, a method for obtaining bicubic spline interpolation surface in three-dimensional space is introduced. To this end, the n×m parametric surface is used as the calculation surface, and the (n+1)×(m+1) scattered points in the space are considered. The multi-power polynomial composed of i and j can be used for solution, and the solution can be solved in the form of a matrix. The spatial grid is formed by connecting two adjacent points in the point set with a line segment. It is used as a control network and can be defined as:

[0063]

[0064] Where Q ij It is the control vertex of the surface. u and v represent the parameters in two directions, i.e. 0<u<1 and 0<v<1; Y in (u) and Y jm (v) represents the spline-based function in the u and v directions on the plane coordinates. When m = n = 3, the characteristic grid of the bicubic B-spline surface is formed, and the equation of the bicubic B-spline surface can be expressed as:

[0065]

[0066] in

[0067] U=[u 3 u 2u 1],V=[v 3 v 2 v 1],

[0068]

[0069] Step S33: The data processing module processes the data received from the depth camera, and uses the pixel-level high-precision information unique to the image to correct the fitting function obtained in step S32, to make up for the lack of precision of the point cloud information, and obtain a panoramic image of the construction. The specific steps are as follows:

[0070] Step S331: using the acquired depth information as a backup and supplement to the point cloud acquired by the 3D laser scanner to correct the fitted surface equation;

[0071] Step S332: Align the acquired image information with the point cloud data of the corresponding position to obtain a more comprehensive modeling.

[0072] Step S34: The data processing module forms a normal vector matrix for the fitting surface. The specific steps are as follows:

[0073] Step S341: The nth fitting surface is denoted as S n =f n (x, y, z), n = 1, 2, 3, ..., on the fitting surface S1 = f1 (x, y, z) of the first monitoring result, a certain number of points a are uniformly selected i,j As the observation point, its coordinates are (x i,j ,y i,j ,z i,j ), a i,j Denoted as matrix A, that is:

[0074]

[0075] Step S342: Calculate each observation point a i,j The normal vector is:

[0076] f i,j =(S′ 1,x (x i,j ,y i,j ,z i,j ),S′ 1,y (x i,j ,y i,j ,z i,j ),S′ 1,z (x i,j ,y i,j ,z i,j ))

[0077] Step S343: f i,jIt is recorded as the normal vector matrix F, that is:

[0078]

[0079] Step S35: The early warning module calculates the tunnel observation point a i,j The specific steps are as follows:

[0080] Observation point a i,j The deformation at the point is the normal vector of the point between the fitting surface S1 = f1 (x, y, z) and S n =f n The modulus between (x,y,z), that is: It can be recorded as the tunnel deformation matrix Δ, that is:

[0081]

[0082] Step S36: The early warning module converts the tunnel deformation value |Δ i,j | is compared with the set threshold Δ0. If the tunnel deformation value |Δ i,j |If it exceeds the threshold Δ0, an early warning is issued.

[0083] Step S4: The intelligent management terminal learns according to the monitoring data, processing results, and early warning conditions, and finally achieves the purpose of prediction. The specific steps are as follows:

[0084] The prediction module uses a CNN neural network to extract the image data obtained in step S33 during the early construction process as features, and puts the features and the corresponding point cloud data obtained in step S3, the corresponding fitting surface obtained in step S34, the corresponding deformation obtained in step S35, and whether a warning is issued in step S36 into the CNN neural network for joint training. In subsequent construction, the training results can be used to predict tunnel deformation based on the obtained monitoring data.

Claims

1. An intelligent monitoring and early warning system for large-section water diversion tunnel construction in red soft rock, characterized in that The system includes an automated integrated monitoring device and an intelligent management terminal, wherein: The automated integrated monitoring device includes a power module, a three-dimensional laser scanning module, a depth camera module, and a data storage and transmission module; The power module is used to carry and transport the three-dimensional laser scanning module, the depth camera module, and the data storage and transmission module, change the position in the water transfer tunnel, and determine the position coordinate information of the collection point; The three-dimensional laser scanning module is used to obtain point cloud data in the tunnel; The depth camera module is used to obtain depth information and image information; The data storage and transmission module is connected to the 3D laser scanning module and the depth camera module, and is used to store the point cloud data collected by the 3D laser scanning module, the depth information and image information obtained by the depth camera module, and the position coordinate information of the collection point, and transmit it to the intelligent management terminal via wireless. The intelligent management terminal includes a data processing module, an early warning module, and a prediction module; The data processing module is connected to the data storage and transmission module and is used to process the data acquired by the three-dimensional laser scanning module and the depth camera module; The early warning module is used to compare the result processed by the data processing module with the set value, and to issue an alarm if the threshold is exceeded; The prediction module is used to learn the data obtained by the three-dimensional laser scanning module and the depth camera module to achieve the purpose of predicting subsequent monitoring data.

2. A method for intelligent monitoring and early warning of large-section water diversion tunnel construction in red soft rock using the system described in claim 1, characterized in that The method comprises the following steps: Step S1: Install the 3D laser scanning module, the depth camera module, and the data storage and transmission module on the platform of the power module for self-inspection and calibration; Step S2: placing the automated integrated monitoring device at the monitoring starting point, and issuing monitoring instructions to the automated integrated monitoring device by the intelligent management terminal to start acquiring point cloud data and image data; Step S3: The intelligent management terminal receives the monitoring data, analyzes and processes it, and issues an early warning in a timely manner according to the processing results; Step S4: The intelligent management terminal learns according to the monitoring data, processing results, and early warning conditions, and ultimately achieves the purpose of prediction.

3. The intelligent monitoring and early warning method for construction of a large-section water conveyance tunnel in red soft rock according to claim 2 is characterized in that The specific steps of step S1 are as follows: Step S11: Install the 3D laser scanning module, the depth camera module, and the data storage and transmission module on the platform of the power module, and check the firmness of the installation; Step S12: Check the environment perception radar of the power module itself to ensure that it can perceive accurately and perform automatic obstacle avoidance and automatic stopping; Step S13: calibrate and level the three-dimensional laser scanning module, the depth camera module and its supporting rotation mechanism and angle detection mechanism.

4. The intelligent monitoring and early warning method for construction of a large-section water conveyance tunnel in red soft rock according to claim 2 is characterized in that The specific steps of step S2 are as follows: Step S21: After the automated integrated monitoring device is inspected, it is transported to the tunnel monitoring starting point and placed according to the location; Step S22: Check and calibrate the 3D laser scanning module and the depth camera module again, and check whether the data storage and transmission modules can work normally and whether the data transmission is disturbed; Step S23: The intelligent management terminal issues a monitoring instruction to the automated integrated monitoring device, and sets key information such as monitoring content, scanning speed, and stop condition; Step S24: The three-dimensional laser scanning module and the depth camera module mounted on the power module continuously acquire the point cloud data and the corresponding depth information and image information inside the tunnel as the vehicle moves forward, and the repeatedly acquired part can be used as data optimization; Step S25: The data storage and transmission module mounted on the power module synchronously transmits the data collected by the three-dimensional laser scanning module and the depth camera module, as well as the camera angle information and the collection point coordinate information to the intelligent management terminal, and backs it up to its own storage module.

5. The intelligent monitoring and early warning method for construction of a large-section water conveyance tunnel in red soft rock according to claim 2 is characterized in that The specific steps of step S3 are as follows: Step S31: the data processing module backs up the received data again; Step S32: the data processing module performs B-spline interpolation fitting on the received point cloud data; Step S33: The data processing module processes the data received from the depth camera, and uses the pixel-level high-precision information unique to the image to correct the fitting function obtained in step S32, so as to make up for the lack of precision of the point cloud information and obtain a panoramic image of the construction; Step S34: the data processing module forms a normal vector matrix for the fitting surface; Step S35: The early warning module calculates the tunnel observation point a i,j Deformation at Step S36: The early warning module converts the tunnel deformation value |Δ i,j | is compared with the set threshold Δ0. If the tunnel deformation value |Δ i,j |If it exceeds the threshold Δ0, an early warning is issued.

6. The intelligent monitoring and early warning method for construction of a large-section water conveyance tunnel in red soft rock according to claim 5 is characterized in that The specific steps of step S33 are as follows: Step S331: using the acquired depth information as a backup and supplement to the point cloud acquired by the 3D laser scanner to correct the fitted surface equation; Step S332: Align the acquired image information with the point cloud data of the corresponding position to obtain a more comprehensive modeling.

7. The intelligent monitoring and early warning method for construction of a large-section water conveyance tunnel in red soft rock according to claim 5 is characterized in that The specific steps of step S34 are as follows: Step S341: The nth fitting surface is denoted as S n =f n (x, y, z), n = 1, 2, 3, ..., on the fitting surface S1 = f1 (x, y, z) of the first monitoring result, a certain number of points a are uniformly selected i,j As the observation point, its coordinates are (x i,j ,y i,j ,z i,j ), a i,j Denoted as matrix A, that is: Step S342: Calculate each observation point a i,j The normal vector is: f i,j =(S1' ,x (x i,j ,y i,j ,z i,j ),S1' ,y (x i,j ,y i,j ,z i,j ),S1' ,z (x i,j ,y i,j ,z i,j )) Step S343: f i,j It is recorded as the normal vector matrix F, that is:

8. The intelligent monitoring and early warning method for construction of a large-section water conveyance tunnel in red soft rock according to claim 7 is characterized in that The specific steps of step S35 are as follows: Observation point a i,j The deformation at the point is the normal vector of the point between the fitting surface S1 = f1 (x, y, z) and S n =f n The modulus between (x,y,z), that is: It can be recorded as the tunnel deformation matrix Δ, that is:

9. The intelligent monitoring and early warning method for construction of a large-section water conveyance tunnel in red soft rock according to claim 5 is characterized in that The specific steps of step S4 are as follows: The prediction module uses a CNN neural network to extract the image data obtained in step S33 during the early construction process as features, and combines the features with the corresponding point cloud data obtained in step S3, the corresponding fitting surface obtained in step S34, the corresponding deformation obtained in step S35, and whether a warning is issued in step S36, and puts them into the CNN neural network for joint training. In subsequent construction, the training results can be used to predict tunnel deformation based on the obtained monitoring data.

Images