A tunnel waterproof sealing structure construction method based on intelligent trolley

Abstract

The present invention belongs to the technical field of tunnel construction, and discloses a construction method of a tunnel waterproof sealing structure based on an intelligent trolley. The precise tunnel centerline position and elevation data are obtained by tunnel section scanning. The intelligent trolley can perform accurate height compensation and path planning based on these data to ensure the accuracy of construction. Process monitoring realizes real-time monitoring of construction quality through image acquisition equipment and image analysis technology. Once a defect is found, the system can be immediately adjusted according to preset rules, reducing dependence on people and reducing the risk of human operational errors. Ensure construction quality. Storing construction data and information through an information management system can not only achieve traceability and transparency of the construction process, but also provide valuable data support for later tunnel monitoring and maintenance.

Description

Technical Field

[0001] The present invention belongs to the technical field of tunnel construction methods, and in particular relates to a tunnel waterproof sealing structure construction method based on an intelligent trolley. Background Art

[0002] The construction of tunnel waterproof sealing structure is an important part of tunnel construction and an important process that affects the quality of tunnel. At present, tunnel waterproof construction trolleys are mostly used for operations. The use of construction trolleys can greatly reduce the labor intensity on site and realize semi-automatic and automated construction. However, the current waterproof construction trolleys have disadvantages such as high equipment cost and incomplete functions. For example, in the construction of waterproof sealing structure, the laying of waterproof materials (waterproof membranes, waterproof boards) is an important process. During the laying process, if the waterproof construction trolley cannot be well matched with the center line of the tunnel, and the laying height is consistent with the height of the tunnel section, it is easy to cause bulging and wrinkling of the waterproof material, which directly affects the waterproof effect. At present, on-site workers are mostly concerned about bulging and wrinkling during the construction process, which increases the workload and is prone to omission of defects. Summary of the invention

[0003] The purpose of the present invention is to propose a tunnel waterproof sealing structure construction method based on an intelligent trolley to solve one or more technical problems existing in the prior art and at least provide a beneficial choice or create conditions.

[0004] In order to achieve the above-mentioned purpose, according to one aspect of the present invention, a method for constructing a tunnel waterproof sealing structure based on an intelligent trolley is provided, and the method comprises the following steps: S100: tunnel section scanning to obtain tunnel section data, including tunnel centerline position and elevation data; S200: preparation and configuration of the intelligent trolley, determining the construction data of the intelligent trolley according to the tunnel centerline position and elevation data, including the walking system path and height compensation data; S300: automated construction, the intelligent trolley automatically advances according to the preset construction path, and compensates the position of the arc track after reaching the laying position, and then scans the laying surface of the waterproof material through tunnel section scanning to obtain the laying surface section data, and compensates the arc track position and the laying trolley position according to the section data, and then performs automated construction; S400: process monitoring, data collection of the laying of waterproof materials is performed through the construction image acquisition equipment, and the construction quality is monitored in real time through data analysis. If defects occur, construction adjustments, suspension of construction or other operations are performed according to preset rules; S500: information management, the scanning data and intelligent trolley working data and construction operation data during the construction process are stored in the information management system.

[0005] Furthermore, the scanning position for the tunnel section in step S100 includes the position of the center line of the waterproof material after the waterproof material is laid.

[0006] Furthermore, the intelligent construction trolley in step S200 includes a walking system, a main frame, a laying trolley assembly, and a laying trolley arc track assembly; the walking system is a track-type or wheeled walking system; the walking system is arranged at the lower part of the main frame; the laying trolley arc track assembly is arranged at the upper part of the main frame, including an arc track and an adjustable track support assembly, and the arc track is connected to the main frame through the adjustable track support assembly so that the position can be adjusted.

[0007] Furthermore, the adjustable track support assembly includes several height adjustment cylinders, connecting seats, several horizontal adjustment cylinders, and a sliding base; the sliding base is slidably arranged on the upper part of the main frame, the bottom of the height adjustment cylinder is fixed on the sliding base, and the top of the piston rod is connected to the connecting seat; the sliding base is connected to the piston rod of the horizontal adjustment cylinder, and the other end of the horizontal adjustment cylinder is fixed on the upper part of the main frame; the laying trolley assembly includes a laying trolley and a scanning trolley, and the laying trolley and the scanning trolley can reciprocate along the arc track, and the scanning trolley is provided with an image acquisition device and a tunnel section scanning device.

[0008] Furthermore, the laying trolley includes a trolley sliding base, a rotating plate, a laying roller connecting frame, and a laying roller; the trolley sliding base can slide back and forth along the arc track, and the rotating plate is arranged on the top surface of the trolley sliding base, and its center is rotatably connected to the center of the top surface of the trolley sliding base; slots are arranged on both sides of the rotating plate, and studs passing through the slots are arranged at corresponding positions on the top surface of the trolley sliding base; the laying roller connecting frame is arranged on the rotating plate, and includes a hinged frame and a support cylinder, one end of the hinged frame is hinged to the rotating plate, and the other end is fixedly connected to the laying roller, and a support cylinder is hinged at the end of the hinged frame connected to the laying roller, and the other end of the support cylinder is hinged to the rotating plate; the support cylinder is a hydraulic cylinder or a gas spring; one end of the rotating plate is connected to the top of the trolley sliding base through a rotating oil cylinder, and both ends of the rotating oil cylinder are hinged to the rotating plate and the top of the trolley sliding base.

[0009] Furthermore, in step S300, the position of the arc track includes a height and / or a horizontal position, and the position of the laying trolley includes a laying roller height and / or an angle compensation.

[0010] Furthermore, the defects in step S400 include bulges and wrinkles; the construction image acquisition device is a combination of any one or more of the scanning vehicle image acquisition device and the tunnel section scanning device, and the image acquisition device is an ultrasonic image acquisition device or a high-definition image acquisition device.

[0011] Furthermore, the image acquisition device on the scanning trolley is a high-definition image acquisition device, and the real-time monitoring of the construction quality through data analysis in step S400 includes the following steps: 1) After the waterproof material is laid, a surface photo of the waterproof material after laying is obtained through the high-definition image acquisition device, and the collected image is pre-processed, including denoising, contrast enhancement, and grayscale conversion; 2) Hough transform is used to detect the characteristics of the ring or elliptical shape, and the bulge risk area is identified; 3) For the pre-processed photo, an image of a preset width of the edge of the waterproof material is obtained, and a grayscale curve of the image along the laying direction is obtained, and the peak and trough change data of the grayscale curve in the preset interval are counted, and if it exceeds the preset threshold, it is marked as a wrinkle risk area; 4) The bulge risk area and the wrinkle risk area are classified according to the preset rules, and the construction adjustment, suspension of construction or other operations are performed according to the classification results and the preset rules; the preset rules are the number and size of the bulge risk area and the wrinkle risk area.

[0012] The beneficial effects of the present invention are as follows: the present invention provides a tunnel waterproof sealing structure construction method based on an intelligent trolley, which can adaptively adjust to match different construction sites to ensure the continuity and sealing of the waterproof layer through the introduction of intelligent control and automation technology, and reduces the possibility of human operational errors through automated construction, thereby increasing the construction speed.

[0013] Specifically, the precise tunnel centerline position and elevation data are obtained through tunnel section scanning. The intelligent trolley can perform accurate height compensation and path planning based on these data to ensure the accuracy of construction. Process monitoring uses image acquisition equipment and image analysis technology to achieve real-time monitoring of construction quality. Once defects are found, the system can immediately adjust according to preset rules, reducing dependence on people and reducing the risk of human operation errors. Ensure construction quality. Storing construction data and information through an information management system can not only achieve traceability and transparency of the construction process, but also provide data support for later tunnel monitoring and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 Shown is a flow chart of a tunnel waterproof sealing structure construction method based on an intelligent trolley.

[0015] Figure 2 Shown is an overall schematic diagram of the intelligent trolley of the present invention.

[0016] Figure 3 Shown is a partially enlarged schematic diagram of the arc track assembly of the intelligent trolley of the present invention.

[0017] Figure 4 Shown is a partially enlarged schematic diagram of the track adjustable support assembly of the intelligent trolley of the present invention.

[0018] Figure 5 Shown is a schematic diagram of a laying trolley of the intelligent trolley of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects of the present invention, so as to fully understand the purpose, scheme and effect of the present invention. It should be noted that the embodiments and features in the embodiments of this application can be combined with each other without conflict.

[0020] like Figure 1 The figure shows a flow chart of a tunnel waterproof sealing structure construction method based on an intelligent trolley. The tunnel waterproof sealing structure construction method based on an intelligent trolley involved in the present invention comprises the following steps:

[0021] S100: tunnel section scanning to obtain tunnel section data, including tunnel centerline position and elevation;

[0022] S200: Preparation and configuration of the intelligent trolley, determining the construction data of the intelligent trolley according to the tunnel centerline position and elevation data, including the travel system path and height compensation data;

[0023] S300: Automated construction: the intelligent trolley automatically moves forward according to the preset construction path, and compensates for the position of the curved track after reaching the laying position. Then, the waterproof material laying surface is scanned through tunnel cross-section scanning to obtain the laying surface cross-section data. The position of the curved track and the laying trolley are compensated according to the cross-section data, and then automated construction is carried out;

[0024] S400: Process monitoring, collects data on the laying of waterproof materials through construction image acquisition equipment, and monitors the construction quality in real time through data analysis. If defects occur, the system will make construction adjustments, suspend construction or perform other operations according to preset rules.

[0025] S500: Information management. The scanning data, smart trolley working data, and construction operation data during the construction process are stored in the information management system to achieve traceability and transparency of the construction process, and can assist in subsequent monitoring and maintenance. The information management system in step 500 is a commonly used product in the field, which is used to store relevant data during the tunnel construction process. The scanning data during the construction process includes the cross-sectional data and waterproof material laying data obtained in steps S100, S300, and S400. The smart trolley working data includes the construction data and position compensation data in steps S200 and S300. The construction operation data includes operation-related data, such as waterproof material type, laying speed and other process data.

[0026] Furthermore, in step S100, the tunnel section scanning can be performed by currently commonly used tunnel section scanning equipment, and the tunnel section point cloud data is obtained by scanning, and the tunnel centerline position and the elevation data on the section are calculated based on the point cloud data.

[0027] Specifically, in step S100, the tunnel section scanning position includes the position of the center line of the waterproof material after the waterproof material is laid. The calculation method of the position of the center line of the waterproof material is as follows: obtain the width l1 of the waterproof material and the overlapping width l2 of the two waterproof materials, thereby calculating the specific scanning position. The scanning positions are respectively l1 / 2, l1 / 2+(l1-2*l2), l1 / 2+2*(l1-2*l2)...l1 / 2+i*(l1-2*l2) from the starting point.

[0028] In addition to the location of the center line of the waterproof material, the sampling density can be further increased to improve accuracy, such as taking the midpoint of the two center lines. However, for the convenience of data collection and processing, an embodiment of the present application only collects tunnel section data at the location of the center line of the waterproof material.

[0029] The process of obtaining tunnel section point cloud data through scanning is generally a built-in function of the sampling equipment. Calculating the tunnel centerline position and elevation data on the section through point cloud data is common knowledge in the field.

[0030] As attached Figure 2-5 As shown, the intelligent construction trolley 1 in step S200 includes a walking system 11, a main frame 12, a laying trolley assembly 13, and a laying trolley arc track assembly 14.

[0031] The walking system 11 can be a track-type or wheel-type walking system, and the specific structure is conventional knowledge in the field. The walking system 11 is arranged at the lower part of the main frame 12.

[0032] The laying trolley circular arc track assembly 14 is arranged on the upper part of the main frame 12, and includes an arc track 141 and an adjustable track support assembly 142. The arc track 141 is connected to the main frame 1 in an adjustable position through the adjustable track support assembly 142. The position and height data of the laying trolley assembly 13 are compensated by the walking system and the adjustable track support assembly 142.

[0033] The track adjustable support assembly 142 includes a plurality of electric cylinders or hydraulic cylinders. Specifically, the track adjustable support assembly 142 includes a plurality of height adjustment cylinders d1, a connecting seat d2, a plurality of level adjustment cylinders d3, and a sliding base d4.

[0034] The sliding base d4 is slidably arranged on the upper part of the main frame 12, the bottom of the height adjustment cylinder d1 is fixed on the sliding base d4, and the top of the piston rod is connected to the connecting base d2. The connecting base d2 is fixedly connected to the arc track 141;

[0035] The sliding base d4 is connected to the piston rod of the horizontal adjustment cylinder d3, and the other end of the horizontal adjustment cylinder d3 is fixed to the upper part of the main frame 12.

[0036] Furthermore, downward protrusions d21 are provided at both ends of the connection seat d2, and a groove d22 is provided on the main frame 12 corresponding to the protrusion d21. The size of the groove is larger than the protrusion d21, and the protrusion d21 extends into the groove d22 to provide safety protection for the connection seat d2.

[0037] The construction path of the intelligent trolley is determined by obtaining the centerline position according to step S100, the height compensation data of the intelligent trolley at the corresponding position is set according to the elevation data, and the extension height of multiple electric cylinders or hydraulic cylinders of the track adjustable support assembly 142 is set according to the height compensation data.

[0038] The laying trolley assembly 13 includes a laying trolley 131 and a scanning trolley 132. The laying trolley 131 and the scanning trolley 132 can reciprocate along the arc track 141. The scanning trolley 132 is provided with an image acquisition device a and a tunnel section scanning device b.

[0039] In step S300, the waterproof material laying surface is scanned by tunnel cross-section scanning, which is performed by the scanning trolley 132. Specifically, when the intelligent trolley drives to a laying position according to the construction path, the elevation data obtained in step S100 is used to compensate the height of the arc track 141, and then the scanning trolley 132 is used to scan the waterproof material laying surface to obtain cross-section data.

[0040] The tunnel section scanning device b of the scanning trolley 132 is used to scan the paving surface of the waterproof material to obtain the paving surface section data. In one embodiment of the present application, the paving surface section data is obtained by using the GOSLAM three-dimensional laser scanning mobile measurement system. In order to reduce costs, multiple laser rangefinders can also be set to collect section data. In one embodiment of the present application, two laser rangefinders are set on both sides of the scanning trolley 132, and more accurate data of the paving surface can be obtained to perform secondary compensation for the arc track 141 and the paving trolley position.

[0041] The laying trolley 131 includes a laying roller c and a compensation unit e, wherein the compensation unit can compensate for the angle and position of the laying roller. Specifically, the laying trolley 131 includes a trolley sliding base e1, a rotating plate e2, a laying roller connecting frame e3, and a laying roller c.

[0042] The trolley sliding base e1 can slide back and forth along the arc track 141, and the rotating plate e2 is arranged on the top surface of the trolley sliding base e1, and its center is rotatably connected to the center of the top surface of the trolley sliding base e1.

[0043] Slots e21 are provided on both sides of the rotating plate e2, and studs passing through the slots e21 are provided at corresponding positions on the top surface of the trolley sliding base e1.

[0044] The laying roller connecting frame e3 is arranged on the rotating plate e2, and includes a hinge frame e31 and a support cylinder e32. One end of the hinge frame e31 is hinged to the rotating plate e2, and the other end is fixedly connected to the laying roller c. A support cylinder e32 is hinged at one end of the hinge frame e31 connected to the laying roller c, and the other end of the support cylinder e32 is hinged to the rotating plate e2. The support cylinder e32 is a hydraulic cylinder that can actively adjust the height of the laying roller c, or it can be a gas spring to provide elastic support for laying. Furthermore, one end of the rotating plate e2 is connected to the top of the trolley sliding base e1 through a rotating oil cylinder e4, and both ends of the rotating oil cylinder e4 are hinged to the rotating plate e2 and the top of the trolley sliding base e1. The laying roller c includes an outer frame and a rotating shaft, and the waterproof coiled material is arranged on the rotating shaft.

[0045] The scanning trolley 132 scans the waterproof material laying surface to obtain the cross-sectional data of the laying surface, and sets the laying process and laying parameters according to the cross-sectional data, which specifically includes the following steps: pre-processing the scanned data, including denoising, data correction, etc., to ensure the accuracy and integrity of the data. According to the pre-processed data, the centerline position and elevation information of the laying surface, and whether there is deflection of the tunnel centerline are obtained, and the position of the arc track 141 and the angle of the laying trolley 131 are compensated according to the above data; the laying trolley 131 performs laying operations according to the preset laying parameters (such as speed).

[0046] The specific position of the arc track 141 includes height and horizontal position, wherein the height is adjusted and compensated by the height adjustment cylinder d1, and the horizontal position of the arc track 141 is compensated by the horizontal adjustment cylinder d3 so that its center line coincides with the center line of the tunnel section. The angle of the laying trolley 131 is specifically the angle of the laying roller c, which is compensated by rotating the cylinder e4 to make it parallel to the center line of the tunnel in the laying surface area.

[0047] Generally speaking, the deflection angle of the tunnel centerline is small, which can be supplemented by the rotation of the laying roller c. At the same time, since the waterproof material will overlap during laying, the overlapping area can cover the impact of the small angle deflection.

[0048] The image acquisition device a on the scanning trolley 132 may be an ultrasonic image acquisition device. In one embodiment of the present application, an ultrasonic defect scanning detector is used to acquire ultrasonic images of the laid waterproof material. Hollow bulges and wrinkles may be identified through analysis of the ultrasonic images.

[0049] The image acquisition device a on the scanning vehicle 132 may be a high-definition image acquisition device, which recognizes bulges and wrinkles through an image recognition algorithm. The specific steps are as follows:

[0050] 1) After the waterproof material is laid, the surface photos of the waterproof material after laying are obtained through high-definition image acquisition equipment, and the collected images are pre-processed, including denoising, contrast enhancement, and grayscale conversion;

[0051] For commonly used white waterproof materials, this application adopts the following collection and pretreatment process:

[0052] 1.1 Use high-definition image acquisition equipment to capture images of the waterproof material after laying. To avoid the influence of light, use fill light for supplementary lighting during the acquisition process.

[0053] 1.2 The collected image is denoised by a Gaussian filter, contrast is enhanced by histogram equalization, and grayscale conversion is performed using the following formula, where R, G, and B are the intensities of the red, green, and blue color channels, respectively: I (x, y) = 0.2R + 0.6G + 0.2B. For white waterproof materials, the green channel is enhanced to optimize the details of the processed image.

[0054] 2) Use Hough transform to detect the features of circular or elliptical shapes and identify the bulge risk areas.

[0055] In one embodiment of the present application, the specific process is:

[0056] 2.1 The preprocessed image is binarized and then an edge detection algorithm, specifically the Canny edge detector, is applied to highlight the edge features in the image.

[0057] 2.2 Matlab was used for Hough transform recognition. The minimum length of the major axis / radius accepted in the Hough algorithm was limited to screen out the first risk area of ​​bulges with a size greater than the preset value. Hough was used to identify arcs that intersected the edge of the image, and the area surrounded by the arcs was marked as the first risk area of ​​bulges, which was an incomplete bulge with an edge.

[0058] 2.3 A bulge recognition model for waterproof materials is obtained by machine learning training. The first bulge risk area obtained in step 2.2 is identified by the bulge recognition model, and the recognition result is output. The area identified as bulge by the bulge recognition model is used as the second bulge risk area.

[0059] 3) For the pre-processed photos, obtain an image of the preset width of the edge of the waterproof material, and obtain the grayscale curve of the image along the laying direction. Count the peak and trough change data of the grayscale curve within the preset interval (a number of pixels along the laying direction). If it exceeds the preset threshold, it is marked as a wrinkle risk area.

[0060] The grayscale curve of the image along the laying direction is obtained as follows: in the image of the preset width at the edge of the waterproof material, multiple sampling windows are set, the sampling window width is a preset number of pixels, and the sampling windows are evenly distributed in the image of the preset width at the edge of the waterproof material perpendicular to the laying direction, thereby obtaining the grayscale curves of the multiple sampling windows,

[0061] The peak and trough change data of the grayscale curve and the threshold judgment include the following process: calculating the first-order derivative of the grayscale curve. If the first-order derivative exceeds the preset value, it is judged that there is a wrinkle defect and marked as the first risk area of ​​the wrinkle. For the grayscale curve of the first risk area of ​​the wrinkle, its second-order derivative is solved, and the zero-crossing point in the second-order derivative is marked as the center position of the wrinkle.

[0062] Machine learning training is used to obtain a wrinkle recognition model for waterproof materials. The wrinkle first risk area is identified using the wrinkle recognition model, and the recognition result is output. The area identified as wrinkles by the wrinkle recognition model is used as the second wrinkle risk area.

[0063] In one embodiment of the present application, using a grayscale curve to identify a wrinkle area includes the following steps:

[0064] 1. Obtain the preprocessed image, identify the edge of the waterproof material through the Canny edge detector or manual annotation, and take an image with a width of 20-30 cm from the edge as the wrinkle recognition image for subsequent wrinkle recognition;

[0065] 2. Sliding window sampling: define the size of the sliding window for sampling, including the width (perpendicular to the laying direction), length (parallel to the laying direction), and step length (parallel to the laying direction) of the window. The step length of the sliding window is 1 / 4-1 / 2 of the length. Multiple sliding windows are evenly distributed in the preset width image of the edge of the waterproof material perpendicular to the laying direction. In this embodiment, there are three sliding windows, which are respectively set at the edges and the center of the two sides of the wrinkle recognition image. The width of the sliding window is 2-5cm and the length is 5-20cm. Multiple continuous sampling window images are obtained by sliding the window;

[0066] 3. Obtain the grayscale curve of the sampling window along the laying direction, take the average of the pixel grayscale values ​​in the direction perpendicular to the window width along the laying direction, and obtain the curve of the average grayscale value along the laying direction as the grayscale curve of the sampling window;

[0067] 4. For the grayscale curve of each sampling window, calculate the first-order derivative according to the following formula: ΔI(x) = I(x+1) -I(x), where ΔI(x) represents the grayscale change rate at pixel point x, and I(x+1) and I(x) represent the grayscale values ​​of two adjacent pixels respectively. If ΔI(x) is greater than the prediction threshold, the sampling window is marked as the first risk area for wrinkles. For the grayscale curve of the wrinkle defect area, solve its second-order derivative according to the following formula, find the zero-crossing point of the second-order derivative, and mark it as the center of the wrinkle: Δ²I(x) = ΔI(x+1) - ΔI(x), where Δ²I(x) represents the curvature of the grayscale curve at pixel point x, and ΔI(x+1) and ΔI(x) represent the grayscale change rate of two adjacent pixels respectively.

[0068] 5. The first risk area of ​​the fold is identified by the fold recognition model, and the recognition result is output. The area identified as a fold by the fold recognition model is used as the second risk area of ​​the fold.

[0069] The training methods for the bulge recognition model and the wrinkle recognition model are as follows: m1. Obtain the preprocessed images and construct a sample set through manual annotation. The sample set includes images with different types of bulges and wrinkles manually annotated. The sample set includes a training set, a validation set, and a test set. m2. Use the training set to train the convolutional neural network (CNN) model, and evaluate the model through the validation set and the test set. The model training is completed when the model recognition error is within the prediction range.

[0070] 6. Classify the bulge second risk area and wrinkle second risk area according to the preset rules, and adjust the construction, suspend the construction or perform other operations according to the classification results and the preset rules. The preset rules may be the number and size of the bulge risk area and the wrinkle risk area. The classification process may use software to compare with the corresponding preset values ​​to obtain the classification results. Furthermore, the classification results may be manually verified. Specifically, in one embodiment of the present application, when the size of the second risk area of ​​bulge (the diameter of the ring or ellipse identified by Hough transform) is greater than the preset value or the number of the second risk areas of connected folds is greater than the preset value (reflecting the size of the folds), the construction is suspended. This situation indicates that a bulge that is too large or continuous large folds have occurred, and the scanning trolley 132 is required to perform a second scan of the paving surface, and the trolley position and the arc track position are adjusted according to the situation; when the size of the second risk area of ​​bulge is within the preset value range and the number of second risk areas of bulge in the preset area is within the preset value range and / or the second risk areas of folds are not connected and the number per unit area is less than the preset value, in this case, a small number of smaller bulges and folds appear, and the laying speed can be reduced and / or the flatness of the waterproof membrane on the laying roller can be checked; when the second risk area of ​​bulge is smaller than the preset value, the second risk areas of folds are not connected and the number per unit area is less than the preset value, it indicates that there are minor defects, and the laying speed can be reduced or no treatment is performed. The classification process adopts a combination of manual and software methods. The specific process is as follows: the alarm method is determined according to the preset rules that the second risk area of ​​bulges and the second risk area of ​​wrinkles meet, including alarm sound and stop operation. The preset rules are divided according to the severity of the defects. When the severity of the defects is relatively minor, an alarm sound or alarm voice reminder is issued. When the defects are serious, the machine can be stopped directly. At the same time, for minor defects, the arc track and the position of the laying trolley can be automatically compensated according to the preset rules when the alarm sound or alarm voice reminder is issued. When the alarm is issued, the defect situation and the recommended response situation should be output for the on-site staff to assist in judgment.

[0071] In another embodiment, the surface point cloud data of the waterproof material can be obtained by the tunnel section scanning device b to identify bulges and wrinkles. The method includes the following steps: after the waterproof material is laid, the surface point cloud data of the waterproof material after laying is obtained by the tunnel section scanning device b, and then the point cloud is processed into a contour map, and the features of the ring or ellipse shape are identified by Hough transformation to identify possible bulges. Wrinkles are identified by analyzing the density of the contour map per unit area. More specifically, an image of a preset width of the edge of the contour map is obtained, and a sliding window is used to sample along the laying direction to obtain the density of the contour map in the window. If it exceeds a preset threshold, it is determined that there are wrinkles.

[0072] In another embodiment, bulges and wrinkles can be identified by combining photos obtained by a high-definition image acquisition device and point cloud data on the surface of the waterproof material obtained by a tunnel section scanning device b and then using a recognition algorithm.

[0073] Although the description of the present invention has been quite detailed and has been described in particular with respect to several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, so as to effectively cover the intended scope of the present invention. In addition, the present invention is described above with the embodiments foreseeable by the inventors, and its purpose is to provide a useful description, and those non-substantial changes to the present invention that are not currently foreseen may still represent equivalent changes of the present invention.

Claims

1. A method for constructing a tunnel waterproof sealing structure based on an intelligent trolley, comprising the following steps: S100: scanning a tunnel section to obtain tunnel section data, including tunnel centerline position and elevation data; S200: preparing and configuring the intelligent trolley, determining the intelligent trolley construction data according to the tunnel centerline position and elevation data, including a walking system path and height compensation data, characterized in that: It also includes S300: automated construction, the intelligent trolley automatically moves forward according to the preset construction path, and compensates for the position of the arc track after reaching the laying position, and then scans the waterproof material laying surface through tunnel section scanning to obtain the laying surface section data, and compensates for the arc track position and the laying trolley position according to the section data, and then performs automated construction; step S400: process monitoring, data collection of the laying of waterproof materials through construction image acquisition equipment, and real-time monitoring of construction quality through data analysis. If defects occur, construction adjustments, suspension of construction or other operations are performed according to preset rules; S500: information management, the scanning data and intelligent trolley working data and construction operation data during the construction process are stored in the information management system; The scanning position of the tunnel section in step S100 includes the position of the center line of the waterproof material after the waterproof material is laid; The intelligent construction trolley in step S200 includes a walking system, a main frame, a laying trolley assembly, and a laying trolley arc track assembly; the walking system is a track-type or wheeled walking system; the walking system is arranged at the lower part of the main frame; the laying trolley arc track assembly is arranged at the upper part of the main frame, including an arc track and an adjustable track support assembly, and the arc track is connected to the main frame through the adjustable track support assembly so that the position can be adjusted; the adjustable track support assembly includes a plurality of height adjustment cylinders, a connecting seat, a plurality of horizontal adjustment cylinders, and a sliding base; the sliding base is slidably arranged on the upper part of the main frame, the bottom of the height adjustment cylinder is fixed on the sliding base, the top of the piston rod is connected to the connecting seat, and the connecting seat is fixedly connected to the arc track; the sliding base is connected to the piston rod of the horizontal adjustment cylinder, and the other end of the horizontal adjustment cylinder is fixed to the upper part of the main frame; the laying trolley assembly includes a laying trolley and a scanning trolley, and the laying trolley and the scanning trolley can reciprocate along the arc track, and the scanning trolley is provided with an image acquisition device and a tunnel section scanning device; In step S300, the position of the arc track includes height and / or horizontal position, and the position of the laying trolley includes height and / or angle compensation of the laying roller; In step S400, the defects include bulges and wrinkles; the construction image acquisition device is a combination of any one or more of the scanning vehicle image acquisition device and the tunnel section scanning device, and the image acquisition device is an ultrasonic image acquisition device or a high-definition image acquisition device; The real-time monitoring of the construction quality through data analysis in step S400 includes the following steps: 1) after the waterproof material is laid, a surface photo of the waterproof material after laying is obtained through a high-definition image acquisition device, and the collected image is pre-processed, including denoising, contrast enhancement, and grayscale conversion; 2) Hough transform is used to detect the features of the ring or ellipse shape to identify the bulge risk area; 3) For the pre-processed photos, obtain an image of the preset width of the edge of the waterproof material, and obtain the grayscale curve of the image along the laying direction, and count the peak and trough change data of the grayscale curve within the preset interval. If it exceeds the preset threshold, it is marked as a wrinkle risk area; 4) Classify the bulge risk areas and wrinkle risk areas according to the preset rules, and make construction adjustments, suspend construction or perform other operations according to the classification results and preset rules; The preset rules are the number and size of bulge risk areas and wrinkle risk areas.

2. The tunnel waterproof sealing structure construction method based on the intelligent trolley according to claim 1 is characterized in that: The laying trolley includes a trolley sliding base, a rotating plate, a laying roller connecting frame, and a laying roller; the trolley sliding base can slide back and forth along the arc track, and the rotating plate is arranged on the top surface of the trolley sliding base, and its center is rotatably connected to the center of the top surface of the trolley sliding base; slots are arranged on both sides of the rotating plate, and studs passing through the slots are arranged at corresponding positions on the top surface of the trolley sliding base; the laying roller connecting frame is arranged on the rotating plate, and includes a hinged frame and a support cylinder, one end of the hinged frame is hinged to the rotating plate, and the other end is fixedly connected to the laying roller, and a support cylinder is hinged at one end of the hinged frame connected to the laying roller, and the other end of the support cylinder is hinged to the rotating plate; the support cylinder is a hydraulic cylinder or a gas spring; one end of the rotating plate is connected to the top of the trolley sliding base through a rotating oil cylinder, and both ends of the rotating oil cylinder are hinged to the rotating plate and the top of the trolley sliding base.

3. The tunnel waterproof sealing structure construction method based on the intelligent trolley according to claim 1 is characterized in that: In step S400, "using Hough transform to detect the features of the ring or ellipse shape and identify the bulge risk area" includes the following steps: using matlab to perform Hough transform identification, and filtering out the bulge first risk area whose size is greater than a preset value by limiting the minimum length of the major axis / radius accepted in the Hough algorithm; Hough is used to identify arcs that intersect with the edge of the image, and the area surrounded by the arc is marked as the first risk area of ​​bulge, which is an incomplete bulge with an edge; Machine learning training is used to obtain a bulge recognition model for waterproof materials. The above steps are used to obtain the first bulge risk area, which is identified using the bulge recognition model and the recognition result is output. The area identified as bulge by the bulge recognition model is used as the second bulge risk area.

4. The method for constructing a tunnel waterproof sealing structure based on an intelligent trolley according to claim 3 is characterized in that: The step of obtaining the grayscale curve of the image along the laying direction in the step S400 of "obtaining an image of the preset width of the edge of the waterproof material for the pre-processed photo, and obtaining the grayscale curve of the image along the laying direction, and counting the peak and trough change data of the grayscale curve in the preset interval, and marking it as a wrinkle risk area if it exceeds a preset threshold" is as follows: in the image of the preset width of the edge of the waterproof material, multiple sampling windows are set, and the sampling window width is a preset number of pixels, which are evenly distributed in the image of the preset width of the edge of the waterproof material along the laying direction perpendicular to the grayscale values ​​of the pixels in the multiple sampling windows, thereby obtaining the grayscale curves of the multiple sampling windows, averaging the grayscale values ​​of the pixels in the multiple sampling windows, and obtaining a curve of the average grayscale value along the laying direction as the grayscale curve; the peak and trough change data of the grayscale curve and the threshold judgment include the following process: calculating the first-order derivative of the grayscale curve, and if the first-order derivative exceeds the preset value, it is judged that there is a wrinkle defect, and it is marked as the first wrinkle risk area; Machine learning training is used to obtain a wrinkle recognition model for waterproof materials. The wrinkle first risk area is identified using the wrinkle recognition model, and the recognition result is output. The area identified as wrinkles by the wrinkle recognition model is used as the second wrinkle risk area.

5. The method for constructing a tunnel waterproof sealing structure based on an intelligent trolley according to claim 4 is characterized in that: In the step S400, "classifying the bulge risk area and the wrinkle risk area according to the preset rules, and adjusting the construction, suspending the construction or performing other operations according to the preset rules based on the classification results; the preset rules are the number and size of the bulge risk area and the wrinkle risk area" includes the following steps: classifying the bulge second risk area and the wrinkle second risk area according to the preset rules, and adjusting the construction, suspending the construction or performing other operations according to the preset rules based on the classification results.

Images