Pipe piece gradient measuring structure and measuring method thereof

Abstract

The invention discloses a duct piece gradient measuring structure and a measuring method thereof, relates to the technical field of tunnel construction, and aims to solve the technical problems that a traditional gradient measuring method is clumsy and long in consumed time, and the measuring precision is greatly influenced by the environment. The measuring assembly comprises supporting frames symmetrically and detachably connected to the inner surface of the segment body, a hollow sliding rod is constructed and connected between the top ends of the supporting frames, a measuring mechanism is installed on the outer edge face of the hollow sliding rod in a sliding fit mode, and the measuring mechanism comprises a bearing frame arranged on the outer edge face of the hollow sliding rod in a sliding fit mode. And a limiting rod is slidably inserted into one end of the bearing frame, a limiting block is formed at the top end of the limiting rod, and a photosensitive plate is connected to the bottom end of the limiting rod. The method has the advantages that the measurement precision is improved, and the requirement for high-precision measurement of the gradient of the duct piece is met.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel construction, and more specifically to a segment slope measurement structure and a measurement method thereof. Background Art

[0002] During the shield construction process, after each ring of excavation is completed, the slope of the segment assembled in the previous ring needs to be measured, and the measurement result needs to be compared with the slope of the tunnel design axis, thereby reflecting the deviation value between the slope of the assembled segment and the slope of the design axis. The degree of deviation plays a guiding role in subsequent shield excavation, segment assembly point selection and other factors.

[0003] At present, the traditional method of measuring the slope of the pipe segment adopts the hanging hammer method. Due to the influence of limited space operation, the previous manual measurement method of measuring the slope of the pipe segment requires the cooperation of two surveyors. One person stands in the assembly area and hangs the top of the thin rope on the hanging hammer to the top of the outer arc surface of the pipe segment. The other person uses a steel ruler to measure the distance between the tip of the lead pendant and the end of the pipe segment. The slope of the pipe segment is calculated according to the measurement results. The traditional slope measurement method is clumsy and time-consuming. The measurement accuracy is greatly affected by the environment, and there are safety risks in the measurement environment of personnel. In view of this, we propose a pipe segment slope measurement structure and its measurement method. Summary of the invention

[0004] The purpose of the present invention is to provide a segment slope measurement structure and a measurement method thereof, so as to solve the technical problems that the traditional slope measurement method is relatively clumsy, time-consuming, and the measurement accuracy is greatly affected by the environment.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a segment slope measurement structure, comprising a measurement component detachably mounted on the inner surface of the segment body;

[0006] The measuring assembly comprises a support frame symmetrically and detachably connected to the inner surface of the segment body, a hollow slide bar is connected between the top ends of the support frame, and a measuring mechanism is slidably mounted on the outer edge of the hollow slide bar;

[0007] The measuring mechanism includes a support bracket slidably arranged on the outer edge surface of the hollow sliding rod, a limit rod is slidably inserted inside one end of the support bracket, the top end structure of the limit rod forms a limit block, and the bottom end structure of the limit rod is connected to a photosensitive plate, the bottom of the photosensitive plate is detachably mounted with a roller structure movably abutting against the inner wall of the pipe segment body, a spring is sleeved on the outer edge surface of the limit rod, and two ends of the spring are respectively fixedly connected to the support bracket and the photosensitive plate, and a laser emitter and a camera are detachably arranged inside the other end of the support bracket.

[0008] The present invention also designs a measuring mechanism, in which the roller structure is fitted to the inner wall of the segment body through the elastic force of the spring, and the measuring mechanism is moved along the axial direction of the segment body through the sliding cooperation of the sliding sleeve and the hollow sliding rod. Light spot images of different sizes can be formed on the surface of the photosensitive plate through the change in the slope of the inner wall of the segment body, thereby realizing multi-point measurement, which is beneficial to improving the measurement accuracy and meeting the demand for high-precision measurement of the segment slope.

[0009] Preferably, the hollow sliding rod is a hollow structure, and a strip groove connected to the interior is formed at the bottom of the outer edge surface of the hollow sliding rod, and the internal structure of the supporting bracket is formed with a sliding sleeve, and the sliding sleeve is slidably mounted on the outer edge surface of the hollow sliding rod, and a groove body connected to the strip groove is formed at the bottom of the outer edge surface of the sliding sleeve.

[0010] Preferably, the top end structure of one of the support frames is connected to a positioning frame, and a processing box is detachably mounted on one side of the support frame on which the positioning frame is provided, the laser emitter and the camera are both connected to a wiring harness that runs along a hollow sliding rod and is connected to the processing box, and the processing box has a built-in image processing module.

[0011] Preferably, both ends of the positioning frame are constructed with positioning ends, a plurality of connecting grooves are distributed on the inner edge surface of the segment body, a threaded connecting hole penetrating the segment body is formed on the inner wall of the connecting groove, and the positioning end abuts against the positioning end arranged in the connecting groove.

[0012] Preferably, the bottom ends of the opposite sides of the symmetrically arranged support frames are threadedly mounted with fixing bolts, and one end of the fixing bolts is screwed into the threaded connection hole.

[0013] Preferably, the roller structure comprises a mounting frame detachably arranged on the bottom surface of the photosensitive plate, a wheel frame is detachably mounted on the bottom of the mounting frame, and a roller is movably arranged inside the wheel frame.

[0014] A method for measuring the slope of a pipe segment comprises the following steps:

[0015] S1. Measurement system layout and preparation;

[0016] Insert the bottom ends of the symmetrically arranged support frames into two connection grooves on the same axial direction of the inner wall of the segment body, and screw the fixing bolts into the threaded connection holes to complete the system installation;

[0017] S2, laser emission and spot image acquisition;

[0018] The laser emitting unit is a laser emitter. When the laser emitting unit is activated, a laser beam is emitted toward the photosensitive plate, and a light spot is formed on the surface of the photosensitive plate.

[0019] The image acquisition unit is a camera, which is started to acquire a light spot image formed on the surface of the photosensitive plate;

[0020] By pushing the support frame, the support frame drives the laser emitter and the camera to move, and a number of measuring points are generated in the axial direction of the segment body. A plurality of light spot images are formed on the photosensitive plate by means of the laser emission unit, and the light spot images formed on the surface of the photosensitive plate are respectively collected by the camera;

[0021] Mark the measurement point and acquisition time, and number them based on the international universal coding standard combined with the measurement information. At the same time, the parameters of focal length, aperture value, exposure time, and sensor temperature are recorded and stored using asymmetric encryption to ensure data security.

[0022] Asymmetric encryption is based on the RSA algorithm, and the encryption formula is: c = m e modn, the decryption formula is: m = c d modn, where m is the plaintext, c is the ciphertext, e and d are the encryption and decryption exponents, and n is the product of two large prime numbers.

[0023] Preferably, the method further comprises the following steps:

[0024] S3, spot image analysis;

[0025] The collected spot image is transmitted to the image analysis module through a lossless compression algorithm. The compression algorithm is based on predictive coding and entropy coding methods. Predictive coding uses a context-based adaptive prediction algorithm and a convolutional neural network training prediction model. Entropy coding uses an adaptive arithmetic coding algorithm. The prediction error formula of predictive coding is: where e i , j is the prediction error, x i , j is the current pixel value, is the predicted value, and the loss function of the convolutional neural network adopts the form of a complex cross entropy loss combined with a regularization term;

[0026] The adaptive arithmetic coding algorithm updates the formula according to the symbol probability interval: L(H)=L(H-1)+P(H-1)×A(H-1), U(H)=L(H-1)+P(H)×A(H-1), where L(H) and U(H) are the lower and upper limits of the coding interval of the current symbol, P(H) is the symbol probability, and A(H) is the coding interval width;

[0027] In the image analysis module, the image is first grayed out using a grayscale conversion method based on spectral weighting, which combines the tube segment spectral reflectance model and the light sensitivity function of the human eye;

[0028] The grayscale conversion formula of spectral weighting is: G ij=∑ k S ij , k ×R k ×W k , where G ij is the gray value, S ij , k is the spectral reflectance value of pixel (i, j) at wavelength k, R k is the sensitivity of the human eye to wavelength k, W k is the weight coefficient;

[0029] The median filter algorithm is used for filtering. The filter kernel size is determined according to the noise. The edge pixels are filled with mirror images. The horizontal edge pixels adjust the replication distance according to the gradient information. The corner points and vertical edge pixels are replicated symmetrically. The boundary condition is judged by a classification algorithm based on pixel neighborhood information. The linear or bicubic interpolation algorithm is used for filling. For the median filter algorithm, when the window size is (2m+1)×(2n+1), the pixel value y after filtering is ij It is the middle value after sorting the pixel values ​​in the window;

[0030] The horizontal edge pixel replication distance adjustment formula is: where d ij is the replication distance, g ij is the gradient value;

[0031] The bicubic interpolation formula is: where a ij is the interpolation coefficient.

[0032] Preferably, an adaptive threshold algorithm is used to determine the spot area, grayscale statistics are calculated by sliding a small window across the image, a weighted average algorithm within the window uses a Gaussian function as the weight, and a boundary detection algorithm based on texture feature analysis is used to avoid the problem of the window crossing the boundary between the spot and the background;

[0033] Adaptive threshold algorithm threshold T ij The calculation formula is: ij =m ij +k×s ij , where m ij is the mean grayscale value in the window, s ij is the standard deviation, k is a constant;

[0034] The weighted average algorithm weight formula is: where ω pq is the weight of the pixel in the window (p, q), σ is the standard deviation of the Gaussian function, and the texture feature analysis is performed by calculating the contrast parameter of the gray level co-occurrence matrix. Among them G x and G y are the horizontal and vertical gradients, the central difference formula Where h is the step length;

[0035] Cubic spline interpolation determines the interpolation function by solving a complex system of linear equations;

[0036] The elements of the directional weighted matrix are determined according to the classification results of the support vector machine algorithm, which finds the optimal classification hyperplane by solving a complex quadratic programming problem;

[0037] The size is calculated based on the pixel size of the edge of the light spot after refinement, and the horizontal direction S x , vertical direction S y , combined with calibration parameters converted to actual size, the error model established by combining multivariate linear regression analysis and principal component analysis is used to compensate for the errors caused by pixel quantization or edge blur, and multiple measurements are processed by weighted average algorithm and least squares statistical average;

[0038] The multivariate linear regression analysis model is: y = β 0 +β 1 x 1 +β 2 x 2 +…β i x i +∈, where y is the error value, x i is the variable that affects the error, β i is the regression coefficient, ∈ is the random error, and principal component analysis performs dimensionality reduction by calculating the eigenvalues ​​and eigenvectors of the covariance matrix;

[0039] Weighted average algorithm weight ω i The calculation formula is: where e i is the error estimate of the ith measurement, and the least squares method finds the optimal solution by minimizing the sum of squared errors.

[0040] Preferably, the method further comprises the following steps:

[0041] S4, slope calculation;

[0042] In the horizontal direction, let the spot size change value between adjacent measurement points be ΔS x =S2 x -S1 x , slope P x =ΔS x / d, when there are several measurement points, calculate ΔS in sequence x , the weights are calculated by weighted average algorithm based on variance analysis of spot size change, and the horizontal slope value is calculated comprehensively after weight normalization;

[0043] The variance analysis formula is: Weighted average algorithm weight wij The calculation formula is: where s ij is the variance of the spot size variation between the i-th and j-th measurement sites, n is the number of samples, is the sample mean, m is the number of measurement site pairs;

[0044] The weight normalization formula is:

[0045] In the vertical direction, the spot size change value between adjacent measurement points is set to ΔS y =S2 y -S1 y , slope P y =ΔS y / d, considering the vertical correction factor determined by the segment material and manufacturing process and the influence of gravity, the vertical slope values ​​of several measuring points are calculated similarly to the horizontal direction;

[0046] The vertical correction factor calculation formula is: C y =f(E y , ρ y , μ y ), where C y is the correction factor, E y is the value related to the elastic modulus in the vertical direction, ρ y is the density-related value, μ y is the friction coefficient related value, and function f is a complex physical model function;

[0047] When considering the influence of gravity, the vertical displacement change caused by gravity is calculated based on the mechanical model of the segment body. The mechanical model is obtained by solving complex elastic mechanical equilibrium equations.

[0048] Segment body comprehensive slope During calculation, the covariance of the horizontal and vertical slope measurement values ​​is calculated through the covariance analysis algorithm, and the vector synthesis weight is adjusted to improve the accuracy of the comprehensive slope calculation;

[0049] The covariance calculation formula of the covariance analysis algorithm is:

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1. The present invention adopts sophisticated algorithms and models in the spot image analysis, from the calibration of the image acquisition unit to the complex prediction and entropy coding technology in the lossless compression algorithm, and then to the filtering, threshold segmentation, edge detection and size calculation of the spot image. In the edge detection, the high-order difference approximation method and cubic spline interpolation are used to improve the accuracy, and multiple factors are considered in the slope calculation. These measures can effectively reduce the measurement error, greatly improve the accuracy of the pipe segment slope measurement results, meet the demand for high-precision measurement of the pipe segment slope, provide reliable data support for subsequent engineering applications, achieve the desired effect, and solve the problems of the traditional slope measurement method being clumsy and time-consuming, and the measurement accuracy being greatly affected by the environment.

[0052] 2. The present invention also designs a measuring mechanism, in which the roller structure is fitted to the inner wall of the segment body through the elastic force of the spring, and the measuring mechanism is moved along the axial direction of the segment body through the sliding cooperation of the sliding sleeve and the hollow sliding rod. Light spot images of different sizes can be formed on the surface of the photosensitive plate through the change in the slope of the inner wall of the segment body, thereby realizing multi-point measurement, which is beneficial to improving the measurement accuracy and meeting the demand for high-precision measurement of the segment slope.

[0053] 3. The present invention also uses asymmetric encryption technology to store the image data and related parameter information collected by the image acquisition unit to ensure that the data is not tampered with or leaked during transmission and storage. At the same time, the measurement site number and acquisition time are marked for each collected spot image, and the data and parameters of each link are recorded during the entire measurement process, so that the entire measurement process has good traceability, which not only facilitates the subsequent analysis and evaluation of the measurement results, but also facilitates troubleshooting and tracing when problems occur, thereby improving the reliability and practicality of the measurement system. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a structural schematic diagram of the present invention;

[0055] Figure 2 It is a structural schematic diagram of the measuring component of the present invention;

[0056] Figure 3 It is a schematic diagram of the structure of the measuring mechanism of the present invention.

[0057] Description of the numbers in the figure:

[0058] 1. Segment body; 2. Connection groove; 3. Threaded connection hole; 4. Measuring assembly; 401. Positioning frame; 402. Positioning end; 403. Support frame; 404. Fixing bolt; 405. Processing box; 406. Hollow slide rod; 5. Measuring mechanism; 501. Support frame; 502. Sleeve; 503. Limit rod; 504. Limit block; 505. Spring; 506. Photosensitive plate; 507. Mounting frame; 508. Wheel frame; 509. Roller; 510. Laser transmitter; 511. Camera. DETAILED DESCRIPTION

[0059] like Figure 1-Figure 3 As shown, the present invention relates to a segment slope measurement structure, comprising a measuring assembly 4 detachably mounted on the inner surface of a segment body 1, the measuring assembly 4 comprising a support frame 403 symmetrically detachably connected to the inner surface of the segment body 1, a hollow slide bar 406 is connected between the top ends of the support frame 403, a measuring mechanism 5 is slidably mounted on the outer edge of the hollow slide bar 406, the measuring mechanism 5 comprises a support bracket 501 slidably mounted on the outer edge of the hollow slide bar 406, one end of the support bracket 501 is internally A limit rod 503 is slidably inserted in the part, the top end of the limit rod 503 is structured to form a limit block 504, and the bottom end of the limit rod 503 is structured to be connected with a photosensitive plate 506, and the bottom of the photosensitive plate 506 is detachably mounted with a roller structure that movably abuts against the inner wall of the pipe segment body 1, and the outer edge surface of the limit rod 503 is sleeved with a spring 505 whose two ends are respectively fixedly connected to the support bracket 501 and the photosensitive plate 506, and the other end of the support bracket 501 is detachably provided with a laser emitter 510 and a camera 511.

[0060] In the embodiment of the present invention, the hollow slide bar 406 is a hollow structure, and the outer edge surface bottom of the hollow slide bar 406 is formed with a strip groove connected to the inside thereof, the internal structure of the support frame 501 is formed with a sliding sleeve 502, the sliding sleeve 502 is slidably mounted on the outer edge surface of the hollow slide bar 406, and the outer edge surface bottom of the sliding sleeve 502 is formed with a groove body connected to the strip groove, the top end structure of one of the support frames 403 is connected with a positioning frame 401, and a processing box 405 is detachably installed on one side of the support frame 403 provided with the positioning frame 401, and the laser emitter 510 and the camera 511 are both connected A wiring harness is connected along the hollow slide rod 406 and connected to the processing box 405. The processing box 405 has an image processing module built in. Both ends of the positioning frame 401 are constructed with positioning ends 402. A plurality of connecting grooves 2 are distributed on the inner edge surface of the tube segment body 1. A threaded connecting hole 3 that passes through the tube segment body 1 is formed on the inner wall of the connecting groove 2. The positioning end 402 abuts against the positioning end 402 arranged in the connecting groove 2. The bottom ends of the opposite sides of the symmetrically arranged support frames 403 are threadedly installed with fixing bolts 404. One end of the fixing bolt 404 is screwed into the threaded connecting hole 3.

[0061] In an embodiment of the present invention, the roller structure includes a mounting frame 507 detachably mounted on the bottom surface of the photosensitive plate 506 , a wheel frame 508 is detachably mounted on the bottom of the mounting frame 507 , and a roller 509 is movably arranged inside the wheel frame 508 .

[0062] A method for measuring the slope of a pipe segment comprises the following steps:

[0063] S1. Measurement system layout and preparation;

[0064] Insert the bottom ends of the symmetrically arranged support frames 403 into the two connection grooves 2 on the same axial direction of the inner wall of the segment body 1, and screw the fixing bolts 404 into the threaded connection holes 3 to complete the system installation;

[0065] S2, laser emission and spot image acquisition;

[0066] The laser emitting unit is a laser emitter 510, which is started to emit a laser beam toward the photosensitive plate 506 and form a light spot on the surface of the photosensitive plate 506;

[0067] The image acquisition unit is a camera 511, which is started to acquire a light spot image formed on the surface of the photosensitive plate 506;

[0068] By pushing the support frame 501, the support frame 501 drives the laser emitter 510 and the camera 511 to move, and a plurality of measuring points are generated in the axial direction of the segment body 1. A plurality of light spot images are formed on the photosensitive plate 506 by means of the laser emission unit, and the light spot images formed on the surface of the photosensitive plate 506 are respectively collected by the camera 511.

[0069] Mark the measurement point and acquisition time, and number them based on the international universal coding standard combined with the measurement information. At the same time, the parameters of focal length, aperture value, exposure time, and sensor temperature are recorded and stored using asymmetric encryption to ensure data security.

[0070] Asymmetric encryption is based on the RSA algorithm, and the encryption formula is: c = m e modn, the decryption formula is: m = c d modn, where m is the plaintext, c is the ciphertext, e and d are the encryption and decryption exponents, and n is the product of two large prime numbers;

[0071] S3, spot image analysis;

[0072] The collected spot image is transmitted to the image analysis module through a lossless compression algorithm. The compression algorithm is based on predictive coding and entropy coding methods. Predictive coding uses a context-based adaptive prediction algorithm and a convolutional neural network training prediction model. Entropy coding uses an adaptive arithmetic coding algorithm. The prediction error formula of predictive coding is: where e i , j is the prediction error, x i , j is the current pixel value, is the predicted value, and the loss function of the convolutional neural network adopts the form of a complex cross entropy loss combined with a regularization term;

[0073] The adaptive arithmetic coding algorithm updates the formula according to the symbol probability interval: L(H)=L(H-1)+P(H-1)×A(H-1), U(H)=L(H-1)+P(H)×A(H-1), where L(H) and U(H) are the lower and upper limits of the coding interval of the current symbol, P(H) is the symbol probability, and A(H) is the coding interval width;

[0074] In the image analysis module, the image is first grayed out using a grayscale conversion method based on spectral weighting, which combines the tube segment spectral reflectance model and the light sensitivity function of the human eye;

[0075] The grayscale conversion formula of spectral weighting is: G ij =∑ k S ij , k ×R k ×W k , where Gij is the gray value, S ij , k is the spectral reflectance value of pixel (i, j) at wavelength k, R k is the sensitivity of the human eye to wavelength k, W k is the weight coefficient;

[0076] The median filter algorithm is used for filtering. The filter kernel size is determined according to the noise. The edge pixels are filled with mirror images. The horizontal edge pixels adjust the replication distance according to the gradient information. The corner points and vertical edge pixels are replicated symmetrically. The boundary condition is judged by a classification algorithm based on pixel neighborhood information. The linear or bicubic interpolation algorithm is used for filling. For the median filter algorithm, when the window size is (2m+1)×(2n+1), the pixel value y after filtering is ij It is the middle value after sorting the pixel values ​​in the window;

[0077] The horizontal edge pixel replication distance adjustment formula is: where d ij is the replication distance, g ij is the gradient value;

[0078] The bicubic interpolation formula is: where a ij is the interpolation coefficient;

[0079] Adaptive threshold algorithm is used to determine the spot area, grayscale statistics are calculated by sliding a small window in the image, and the weighted average algorithm in the window adopts Gaussian function weights. At the same time, the boundary detection algorithm based on texture feature analysis is used to avoid the problem of the window crossing the boundary between the spot and the background.

[0080] Adaptive threshold algorithm threshold T ij The calculation formula is: ij =m ij +k×s ij , where m ij is the mean grayscale value in the window, s ij is the standard deviation, k is a constant;

[0081] The weighted average algorithm weight formula is: where ω pq is the weight of the pixel in the window (p, q), σ is the standard deviation of the Gaussian function, and the texture feature analysis is performed by calculating the contrast parameter of the gray level co-occurrence matrix. Among them G x and G y are the horizontal and vertical gradients, the central difference formula Where h is the step length;

[0082] Cubic spline interpolation determines the interpolation function by solving a complex system of linear equations;

[0083] The elements of the directional weighted matrix are determined according to the classification results of the support vector machine algorithm, which finds the optimal classification hyperplane by solving a complex quadratic programming problem;

[0084] The size is calculated based on the pixel size of the edge of the light spot after refinement, and the horizontal direction S x , vertical direction S y , combined with calibration parameters converted to actual size, the error model established by combining multivariate linear regression analysis and principal component analysis is used to compensate for the errors caused by pixel quantization or edge blur, and multiple measurements are processed by weighted average algorithm and least squares statistical average;

[0085] The multivariate linear regression analysis model is: y = β 0 +β 1 x 1 +β 2 x 2 +…β i x i +∈, where y is the error value, x i is the variable that affects the error, β i is the regression coefficient, ∈ is the random error, and principal component analysis performs dimensionality reduction by calculating the eigenvalues ​​and eigenvectors of the covariance matrix;

[0086] Weighted average algorithm weight ω i The calculation formula is: where e i is the error estimate of the ith measurement, and the least squares method finds the optimal solution by minimizing the sum of squared errors;

[0087] S4, slope calculation;

[0088] In the horizontal direction, let the spot size change value between adjacent measurement points be ΔS x =S2 x -S1 x , slope P x =ΔS x / d, when there are several measurement points, calculate ΔS in sequence x , the weights are calculated by weighted average algorithm based on variance analysis of spot size change, and the horizontal slope value is calculated comprehensively after weight normalization;

[0089] The variance analysis formula is: Weighted average algorithm weight w ij The calculation formula is: where s ij is the variance of the spot size variation between the i-th and j-th measurement sites, n is the number of samples, is the sample mean, m is the number of measurement site pairs;

[0090] The weight normalization formula is:

[0091] In the vertical direction, the spot size change value between adjacent measurement points is set to ΔS y =S2 y -S1 y , slope P y =ΔS y / d, considering the vertical correction factor determined by the segment material and manufacturing process and the influence of gravity, the vertical slope values ​​of several measuring points are calculated similarly to the horizontal direction;

[0092] The vertical correction factor calculation formula is: C y =f(E y , ρ y , μ y ), where C y is the correction factor, E y is the value related to the elastic modulus in the vertical direction, ρ y is the density-related value, μ y is the friction coefficient related value, and function f is a complex physical model function;

[0093] When considering the influence of gravity, the vertical displacement change caused by gravity is calculated according to the mechanical model of the segment body 1, and the mechanical model is obtained by solving complex elastic mechanical equilibrium equations;

[0094] Segment body 1 comprehensive slope During calculation, the covariance of the horizontal and vertical slope measurement values ​​is calculated through the covariance analysis algorithm, and the vector synthesis weight is adjusted to improve the accuracy of the comprehensive slope calculation;

[0095] The covariance calculation formula of the covariance analysis algorithm is:

[0096] Working principle: This embodiment provides a segment slope measurement structure and a measurement method thereof. When in use, the bottom ends of the symmetrically arranged support frames 403 are respectively arranged in two connection grooves 2 on the same axis of the inner wall of the segment body 1, and then the fixing bolts 404 are screwed into the threaded connection holes 3 to complete the installation of the measurement assembly 4. Then, the laser transmitter 510 is started to emit laser to the photosensitive plate 506, and a light spot is formed on the surface of the photosensitive plate 506. By starting the camera 511, the light spot image is collected by the camera 511, and then the sliding sleeve 502 and the hollow sliding rod 40 are used to measure the light spot image. The sliding fit of 6 enables the measuring mechanism 5 to move along the axial direction of the segment body 1. The elastic force of the spring 505 makes the roller structure fit the inner wall of the segment body 1, so that light spot images of different sizes can be formed on the surface of the photosensitive plate 506 through the change of the slope of the inner wall of the segment body 1, and a plurality of light spot images are collected through the camera 511. Then, the light spot images are processed and calculated through the image processing module in the processing box 405 to obtain the slope of the segment body 1, thereby realizing the measurement of the slope. The analysis and calculation of multiple sets of data are helpful to improve the accuracy of the slope measurement.

[0097] The embodiments of the present invention disclose preferred embodiments, but are not limited thereto. A person skilled in the art can easily understand the spirit of the present invention based on the above embodiments and make different extensions and changes. However, as long as they do not deviate from the spirit of the present invention, they are all within the protection scope of the present invention.

Claims

1. A segment slope measurement structure, characterized in that: It comprises a measuring component (4) which is detachably mounted on the inner surface of the segment body (1); The measuring assembly (4) comprises a support frame (403) symmetrically and detachably connected to the inner surface of the segment body (1); a hollow sliding rod (406) is connected between the top ends of the support frame (403); and a measuring mechanism (5) is slidably mounted on the outer edge surface of the hollow sliding rod (406); The measuring mechanism (5) comprises a support frame (501) slidably arranged on the outer edge surface of the hollow sliding rod (406); a limit rod (503) is slidably inserted into one end of the support frame (501); the top end of the limit rod (503) forms a limit block (504); the bottom end of the limit rod (503) is connected to a photosensitive plate (506); a roller structure movably abutting against the inner wall of the pipe segment body (1) is detachably installed at the bottom of the photosensitive plate (506); a spring (505) is sleeved on the outer edge surface of the limit rod (503) with two ends respectively fixedly connected to the support frame (501) and the photosensitive plate (506); and a laser emitter (510) and a camera (511) are detachably arranged inside the other end of the support frame (501).

2. A segment slope measurement structure according to claim 1, characterized in that: The hollow slide bar (406) is a hollow structure, and a strip groove connected to the interior is formed at the bottom of the outer edge surface of the hollow slide bar (406). The internal structure of the support frame (501) is formed with a sliding sleeve (502), and the sliding sleeve (502) is slidably mounted on the outer edge surface of the hollow slide bar (406), and a groove body connected to the strip groove is formed at the bottom of the outer edge surface of the sliding sleeve (502).

3. A segment slope measurement structure according to claim 2, characterized in that: The top end structure of one of the support frames (403) is connected to a positioning frame (401), and a processing box (405) is detachably mounted on one side of the support frame (403) provided with the positioning frame (401), the laser emitter (510) and the camera (511) are both connected to a wiring harness that runs along a hollow sliding rod (406) and is connected to the processing box (405), and the processing box (405) has a built-in image processing module.

4. A segment slope measurement structure according to claim 3, characterized in that: Both ends of the positioning frame (401) are structured to form positioning ends (402), a plurality of connection grooves (2) are distributed on the inner edge surface of the segment body (1), a threaded connection hole (3) penetrating the segment body (1) is formed on the inner wall of the connection groove (2), and the positioning end (402) abuts against the positioning end (402) arranged in the connection groove (2).

5. A segment slope measurement structure according to claim 4, characterized in that: The bottom ends of the opposite sides of the symmetrically arranged support frames (403) are threadedly mounted with fixing bolts (404), and one end of the fixing bolts (404) is screwed into the threaded connection holes (3).

6. A segment slope measurement structure according to claim 5, characterized in that: The roller structure comprises a mounting frame (507) detachably arranged on the bottom surface of the photosensitive plate (506), a wheel frame (508) detachably mounted on the bottom of the mounting frame (507), and a roller (509) movably arranged inside the wheel frame (508).

7. A method for measuring a pipe segment slope, comprising a pipe segment slope measuring structure according to claim 6, characterized in that: The following steps are involved: S1. Measurement system layout and preparation; The bottom ends of the symmetrically arranged support frames (403) are respectively inserted into two connection grooves (2) on the inner wall of the segment body (1) in the same axial direction, and the fixing bolts (404) are screwed into the threaded connection holes (3) to complete the system installation; S2, laser emission and spot image acquisition; The laser emitting unit is a laser emitter (510), and the laser emitting unit is started to emit a laser beam toward the photosensitive plate (506), and a light spot is formed on the surface of the photosensitive plate (506); The image acquisition unit is a camera (511), which is started to acquire a light spot image formed on the surface of the photosensitive plate (506); By pushing the support frame (501), the support frame (501) drives the laser emitter (510) and the camera (511) to move, thereby generating a plurality of measurement points in the axial direction of the segment body (1), forming a plurality of light spot images on the photosensitive plate (506) by means of the laser emission unit, and collecting the light spot images formed on the surface of the photosensitive plate (506) respectively by means of the camera (511); Mark the measurement point and acquisition time, and number them based on the international universal coding standard combined with the measurement information. At the same time, the parameters of focal length, aperture value, exposure time, and sensor temperature are recorded and stored using asymmetric encryption to ensure data security. Asymmetric encryption is based on the RSA algorithm, and the encryption formula is: c = m e modn, the decryption formula is: m = c d modn, where m is the plaintext, c is the ciphertext, e and d are the encryption and decryption exponents, and n is the product of two large prime numbers.

8. A segment slope measurement method according to claim 7, characterized in that: The following steps are also included: S3, spot image analysis; The collected spot image is transmitted to the image analysis module through a lossless compression algorithm. The compression algorithm is based on predictive coding and entropy coding methods. Predictive coding uses a context-based adaptive prediction algorithm and a convolutional neural network training prediction model. Entropy coding uses an adaptive arithmetic coding algorithm. The prediction error formula of predictive coding is: where e i , j is the prediction error, x i , j is the current pixel value, is the predicted value, and the loss function of the convolutional neural network adopts the form of a complex cross entropy loss combined with a regularization term; The adaptive arithmetic coding algorithm updates the formula according to the symbol probability interval: L(H) = L(H-1) + P(H-1) × A(H-1), U(H) = L(H-1) + P(H) × A(H-1), where L(H) and U(H) are the lower and upper limits of the coding interval of the current symbol, P(H) is the symbol probability, and A(H) is the coding interval width; In the image analysis module, the image is first grayed out using a grayscale conversion method based on spectral weighting, which combines the tube segment spectral reflectance model and the light sensitivity function of the human eye; The grayscale conversion formula of spectral weighting is: G ij =∑ k S ij,k ×R k ×W k , where G ij is the gray value, S ij,k is the spectral reflectance value of pixel (i, j) at wavelength k, R k is the sensitivity of the human eye to wavelength k, W k is the weight coefficient; The median filter algorithm is used for filtering. The filter kernel size is determined according to the noise. The edge pixels are filled with mirror images. The horizontal edge pixels adjust the replication distance according to the gradient information. The corner points and vertical edge pixels are replicated symmetrically. The boundary condition is judged by a classification algorithm based on pixel neighborhood information. The linear or bicubic interpolation algorithm is used for filling. For the median filter algorithm, when the window size is (2m+1)×(2n+1), the pixel value y after filtering is ij It is the middle value after sorting the pixel values ​​in the window; The horizontal edge pixel replication distance adjustment formula is: where d ij is the replication distance, g ij is the gradient value; The bicubic interpolation formula is: Among them, a ij is the interpolation coefficient.

9. A method for measuring the slope of a pipe segment according to claim 8, characterized in that: It also includes using an adaptive threshold algorithm to determine the spot area, calculating grayscale statistics by sliding a small window across the image, using a weighted average algorithm within the window in the form of a Gaussian function, and using a boundary detection algorithm based on texture feature analysis to avoid the problem of the window crossing the boundary between the spot and the background; Adaptive threshold algorithm threshold T ij The calculation formula is: ij =m ij +k×s ij , where m ij is the mean grayscale value in the window, s ij is the standard deviation, k is a constant; The weighted average algorithm weight formula is: where ω pq is the weight of the pixel in the window (p, q), σ is the standard deviation of the Gaussian function, and the texture feature analysis is performed by calculating the contrast parameter of the gray level co-occurrence matrix. Among them G x and G y are the horizontal and vertical gradients, the central difference formula Where h is the step length; Cubic spline interpolation determines the interpolation function by solving a complex system of linear equations; The elements of the directional weighted matrix are determined according to the classification results of the support vector machine algorithm, which finds the optimal classification hyperplane by solving a complex quadratic programming problem; The size is calculated based on the pixel size of the edge of the light spot after refinement, and the horizontal direction S x , vertical direction S y , combined with calibration parameters converted to actual size, the error model established by combining multivariate linear regression analysis and principal component analysis is used to compensate for the errors caused by pixel quantization or edge blur, and multiple measurements are processed by weighted average algorithm and least squares statistical average; The multivariate linear regression analysis model is: y = β0 + β1x1 + β2x2 + ... β i x i +∈, where y is the error value, x i is the variable that affects the error, β i is the regression coefficient, ∈ is the random error, and principal component analysis performs dimensionality reduction by calculating the eigenvalues ​​and eigenvectors of the covariance matrix; Weighted average algorithm weight ω i The calculation formula is: where e i is the error estimate of the ith measurement, and the least squares method finds the optimal solution by minimizing the sum of squared errors.

10. A method for measuring the slope of a pipe segment according to claim 7, characterized in that: The following steps are also included: S4, slope calculation; In the horizontal direction, let the spot size change value between adjacent measurement points be ΔS x =S2 x -S1 x , slope P x =ΔS x / d, when there are several measurement points, calculate ΔS in sequence x , the weights are calculated by weighted average algorithm based on variance analysis of spot size change, and the horizontal slope value is calculated comprehensively after weight normalization; The variance analysis formula is: Weighted average algorithm weight w ij The calculation formula is: where s ij is the variance of the spot size variation between the i-th and j-th measurement sites, n is the number of samples, is the sample mean, m is the number of measurement site pairs; The weight normalization formula is: In the vertical direction, the spot size change value between adjacent measurement points is set to ΔS y =S2 y -S1 y , slope P y =ΔS y / d, considering the vertical correction factor determined by the segment material and manufacturing process and the influence of gravity, the vertical slope values ​​of several measuring points are calculated similarly to the horizontal direction; The vertical correction factor calculation formula is: C y =f(E y ,ρ y ,μ y ), where C y is the correction factor, E y is the value related to the elastic modulus in the vertical direction, ρ y is the density-related value, μ y is the friction coefficient related value, and function f is a complex physical model function; When the influence of gravity is considered, the vertical displacement change caused by gravity is calculated based on the mechanical model of the segment body (1), which is obtained by solving complex elastic mechanical equilibrium equations; Segment body (1) Comprehensive slope During calculation, the covariance of the horizontal and vertical slope measurement values ​​is calculated through the covariance analysis algorithm, and the vector synthesis weight is adjusted to improve the accuracy of the comprehensive slope calculation; The covariance calculation formula of the covariance analysis algorithm is:

Images