Shield tunneling machine pose measuring system and measuring method

Abstract

The invention relates to the technical field of tunnel construction, in particular to a shield tunneling machine pose measurement system and method, and the system comprises a shield tunneling machine data acquisition module, a pose parameter acquisition module, a pose deviation analysis module, a corrected pose parameter acquisition module, a pose instruction generation module, and a management database. According to the system, three-dimensional point cloud data are input into a pose fusion model to obtain initial pose parameters, matching is performed by defining the range of the shield tunneling machine in the three-dimensional geologic model and screening related points of all axes, pose deviation parameters of the shield tunneling machine are evaluated by calculating position deviation and pose deviation, and the pose fusion model is obtained. The initial pose parameter and the pose deviation parameter are input into the adaptive filtering algorithm for iterative optimization, the corrected pose parameter is obtained, whether the corrected pose parameter exceeds the dynamic threshold range or not is judged, the pose adjustment instruction can be generated in time, and the construction problem caused by too large pose deviation is prevented.

Description

Technical Field

[0001] The invention relates to the technical field of tunnel construction, and in particular to a shield machine posture measurement system and a measurement method. Background Art

[0002] In modern tunnel construction, the shield method has become the mainstream construction method due to its advantages such as high efficiency, safety and little impact on the surrounding environment. The shield machine operates in a complex underground environment. The precise measurement of its position plays a decisive role in ensuring that the tunnel is accurately advanced according to the designed route and ensuring the quality and safety of construction. Accurately grasping the position and posture of the shield machine can effectively avoid tunnel deviation, reduce construction risks and improve the overall benefits of the project. It is a key technical link in the tunnel construction process.

[0003] At present, although there are many kinds of technologies used for measuring the posture of shield machines, they all have certain limitations. For example, the existing Chinese patent with application number 202010649631.1 discloses a vibration-resistant laser target, a shield machine posture measurement system and a measurement method. The scheme collects first angle data of the shield machine at a frequency greater than or equal to a first preset frequency through an angle acquisition unit, and filters the first angle data based on a filtering algorithm to generate second angle data, and determines the posture of the shield machine based on the second angle data and the total station data collected by the total station. This can reduce the influence of the vibration of the shield machine on the measurement results during the posture measurement of the shield machine and improve the measurement accuracy of the posture of the shield machine.

[0004] However, the above patent has the following problems: the solution directly measures the three horizontal parameters of the shield machine through the total station, determines the roll angle and pitch angle based on the second angle data, and then determines the azimuth angle. This method relies on the reliability of angle and simple coordinate measurement. At the same time, when determining the roll angle and pitch angle, it assumes the stability of the reference plane and the absolute accuracy of the sensor installation based on the angle relationship between the axis of the angle sensor coordinate system and the reference plane, which may lead to deviations in the determination of posture parameters in actual measurement.

[0005] For example, the existing Chinese patent with application number 202211579395.6 discloses a shield tunneling posture measurement method and system. This scheme is based on the shield tunneling posture measurement method that integrates a gyroscope and an inclinometer. According to the high-precision data of the inclinometer, the offset angular rate of the gyroscope is accurately estimated and compensated in real time when the shield machine is working in a steady state, thereby improving the working accuracy of the gyroscope. When the inclinometer is subjected to severe vibration, the compensated gyroscope angle is fused with the inclinometer data to obtain the real-time attitude angle information of the shield machine.

[0006] However, the above patent has the following problems: the solution mainly relies on data collected by gyroscopes and inclinometers, and only uses information obtained from these two sensors to measure posture. It cannot fully consider the complex spatial environment around the shield machine, and lacks the comprehensiveness and accuracy of the initial posture parameter acquisition. The accuracy and reliability of posture determination are relatively low. At the same time, the geological environment factors are not included in the posture evaluation system. Under complex geological conditions, it may not be possible to accurately evaluate the actual posture deviation of the shield machine, and the accuracy of the evaluation results is limited. Summary of the invention

[0007] In order to overcome the shortcomings of the background technology, the embodiments of the present invention provide a shield machine posture measurement system and measurement method, which can effectively solve the problems involved in the above-mentioned background technology.

[0008] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides a shield machine posture measurement system, including: a shield machine data acquisition module, used to obtain three-dimensional point cloud data during the shield machine excavation process, the three-dimensional point cloud data includes the real-time three-dimensional coordinates of the shield machine and three-dimensional geological point cloud data.

[0009] The pose parameter acquisition module is used to preprocess the three-dimensional point cloud data and input it into the pose fusion model to obtain the initial pose parameters.

[0010] The posture deviation analysis module is used to define the location of the shield machine in the three-dimensional geological model and select the relevant points of each axis for matching. The posture deviation parameters of the shield machine are evaluated by calculating the position deviation and posture deviation.

[0011] The corrected posture parameter acquisition module is used to iteratively optimize the initial posture parameters and posture deviation parameters to obtain the corrected posture parameters.

[0012] The posture instruction generation module is used to determine whether the corrected posture parameters exceed the dynamic threshold range. If so, a posture adjustment instruction is generated. Otherwise, trend analysis is performed on it together with historical data to output real-time posture warning information.

[0013] The management database is used to store the three-dimensional point cloud data, initial posture parameters, posture deviation parameters, corrected posture parameters and historical data of the shield machine body.

[0014] Preferably, the specific analysis method for preprocessing the three-dimensional point cloud data is: collecting the three-dimensional point cloud data of the shield machine respectively according to the sampling frequency set by the corresponding sensor and recording the corresponding timestamp information, and adjusting the three-dimensional point cloud data of the shield machine to the same time point.

[0015] Preferably, the specific analysis method of the posture parameter acquisition module is: select the Kalman filter model as the posture fusion model, input the three-dimensional point cloud data of the shield machine to obtain the state estimation value of the shield machine, set the Kalman filter model iteration termination condition as the change of the posture parameter is less than the preset threshold, enter the iterative calculation according to the three-dimensional point cloud data of the shield machine and each state estimation value, and after the iterative calculation is completed, extract the three-dimensional position coordinates and attitude angle information of the shield machine from the state estimation value finally output by the model as the initial posture parameters.

[0016] Preferably, the specific analysis method for constructing the three-dimensional geological model is: obtaining the three-dimensional geological point cloud data of the shield machine, clustering the point cloud data belonging to the same tunnel area into one category according to the regional division information of the surrounding tunnels, and extracting the point cloud data corresponding to each cluster.

[0017] Each geometric surface is fitted according to the point cloud data corresponding to each cluster, and they are combined to obtain a complete three-dimensional geological model.

[0018] Preferably, the specific analysis method of the axis-related points is: obtain the front endpoint coordinates, rear endpoint coordinates and radius range of the shield machine from the real-time three-dimensional coordinates of the shield machine, take the shield machine axis as the center, and construct a cylinder in combination with the shield machine radius to define the range of the shield machine in the three-dimensional geological model.

[0019] Along the tunneling direction of the shield machine, the shield machine within the range is divided into cross sections according to the principle of equal spacing to obtain various cross sections, and each of the cross sections is perpendicular to the tunneling direction of the shield machine.

[0020] For each cross section, points located in the cross section are extracted from the three-dimensional geological point cloud data, and their distances to the center of the cross section are calculated respectively, wherein the center of the cross section is the projection point of the shield machine design axis on the cross section.

[0021] A reasonable deviation value is set, and the distance threshold is determined by summing it with the radius of the shield machine. The distance from each point to the center of the cross section is compared with the distance threshold, and the points whose distance to the center of the cross section is less than or equal to the distance threshold are screened out as the relevant points of each axis.

[0022] Linear fitting is performed on the axis-related points until the sum of the squares of the distances from the axis-related points to the fitting straight line reaches a minimum, thereby determining the preliminary direction of the shield machine's excavation axis.

[0023] Preferably, the specific analysis method of the posture deviation parameters is: randomly select two points on the excavation axis, obtain the direction vector of the excavation axis by calculating the vector difference formed by the two points, extract the axis-related points on the shield machine excavation axis, and simultaneously obtain the points on the preset design axis, find the nearest point on the design axis for each axis-related point, establish a corresponding relationship between the two, and form matching point pairs.

[0024] The position difference of each matching point pair in three-dimensional space is calculated respectively, and the position deviation is obtained by taking the average value. The attitude deviation is obtained by calculating the angle between the direction vector of the design axis and the shield machine excavation axis.

[0025] The posture deviation parameters of the shield machine are evaluated based on position deviation and posture deviation.

[0026] Preferably, the specific analysis method of the modified posture parameter acquisition module is: setting the state transfer matrix and the measurement matrix, taking the initial posture parameters of the shield machine as the initial value of the state vector, and predicting the state vector at the next moment by performing matrix multiplication operation on the initial value of the state vector and the state transfer matrix.

[0027] The posture deviation parameters are used as measurement values, and the predicted state vector is updated through the measurement matrix.

[0028] The number of loop executions is determined according to the accuracy requirements of the shield machine's posture parameters. After iterative operations with the set number of loop executions, the obtained state vector is recorded as the corrected posture parameter of the shield machine.

[0029] Preferably, the specific analysis method of the posture instruction generation module is: setting a dynamic threshold range, comparing and judging the acquired corrected posture parameters with the dynamic threshold range, and if the corrected posture parameters exceed the dynamic threshold range, generating a posture adjustment instruction, wherein the posture adjustment instruction takes the direction exceeding the range as the positive direction, and the adjustment direction is to adjust in the negative direction, and determines the adjustment amplitude according to the exceeding range, and sends the posture adjustment instruction to the system to execute the corresponding posture adjustment operation.

[0030] If the corrected posture parameter does not exceed the dynamic threshold range, the corrected posture parameter is subjected to trend analysis with historical data, where the historical data is posture angle data accumulated during the previous operation of the shield machine.

[0031] Preferably, the posture instruction generation module also includes: selecting a set number of time points as the time window size, organizing the historical data into a historical data sequence in time point order, starting from the first data point of the historical data sequence, and taking the time window as a unit, calculating the average value of the posture angle data in each time window to obtain a moving average value sequence of the historical data sequence.

[0032] Perform linear regression on the moving average sequence and calculate the slope of the regression line. If the slope of the regression line is positive, it means that the attitude angle of the shield machine is generally on an upward trend. If the slope of the regression line is negative, it means that the attitude angle of the shield machine is generally on a downward trend. If the slope of the regression line is within the set slope threshold range, it means that the stability of the attitude angle of the shield machine is qualified. The change of the attitude angle of the shield machine is output to the system as real-time posture warning information.

[0033] Preferably, the present invention provides a method for measuring the position and posture of a shield machine, and the specific steps of the measurement method are as follows: S1. Shield machine data acquisition: Acquire three-dimensional point cloud data of the shield machine during excavation, and the three-dimensional point cloud data includes the real-time three-dimensional coordinates of the shield machine and three-dimensional geological point cloud data.

[0034] S2. Acquisition of pose parameters: Preprocess the 3D point cloud data and input it into the pose fusion model to obtain the initial pose parameters.

[0035] S3. Posture deviation analysis: define the location of the shield machine in the three-dimensional geological model and select the relevant points of each axis for matching, and evaluate the posture deviation parameters of the shield machine by calculating the position deviation and posture deviation.

[0036] S4. Obtaining corrected posture parameters: Iteratively optimize the initial posture parameters and posture deviation parameters to obtain corrected posture parameters.

[0037] S5. Posture instruction generation: Determine whether the corrected posture parameters exceed the dynamic threshold range. If so, generate a posture adjustment instruction. Otherwise, perform trend analysis on it together with historical data and output real-time posture warning information.

[0038] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: 1. The present invention obtains initial pose parameters by inputting three-dimensional point cloud data into the pose fusion model, and can make more comprehensive use of spatial information. The three-dimensional point cloud data can provide detailed geometric information of the environment around the shield machine. The pose fusion model can integrate this information for more accurate pose estimation, thereby improving the accuracy and reliability of pose determination.

[0039] 2. The present invention defines the scope of the shield machine in the three-dimensional geological model and selects the relevant points of each axis for matching. It evaluates the posture deviation parameters of the shield machine by calculating the position deviation and posture deviation. It can closely combine the geological environment of the actual construction of the shield machine to evaluate the posture, and can better reflect the actual posture of the shield machine under complex geological conditions, making the evaluation result more accurate.

[0040] 3. The present invention obtains corrected posture parameters by inputting the initial posture parameters and posture deviation parameters into an adaptive filtering algorithm for iterative optimization, and determines whether the corrected posture parameters exceed the dynamic threshold range. If so, a posture adjustment instruction is generated. Otherwise, a trend analysis is performed with the historical data to output real-time posture warning information. Through dynamic threshold judgment, posture adjustment instructions can be generated in time to prevent construction problems caused by excessive posture deviation and ensure construction safety and accuracy. When the threshold is not exceeded, a trend analysis is performed with the historical data to output real-time warning information, so that the posture change trend can be known in advance to prevent potential risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0042] Figure 1 This is a module connection diagram of a shield machine posture measurement system.

[0043] Figure 2 for Figure 1 Flowchart of the relevant points of each axis in the mid-pose deviation analysis module.

[0044] Figure 3 The figure is a flow chart of a shield machine posture measurement method. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] See also Figure 1 As shown, a shield machine posture measurement system includes a shield machine data acquisition module, a posture parameter acquisition module, a posture deviation analysis module, a corrected posture parameter acquisition module, a posture instruction generation module, and a management database.

[0047] The management database is connected to the shield machine data acquisition module, the posture parameter acquisition module, the posture deviation analysis module, the corrected posture parameter acquisition module, and the posture instruction generation module; the corrected posture parameter acquisition module is connected to the posture parameter acquisition module, the posture deviation analysis module, and the posture instruction generation module; the posture parameter acquisition module is connected to the shield machine data acquisition module and the posture deviation analysis module.

[0048] The shield machine data acquisition module is used to obtain the three-dimensional point cloud data of the shield machine during the excavation process. The three-dimensional point cloud data includes the real-time three-dimensional coordinates of the shield machine and three-dimensional geological point cloud data.

[0049] The pose parameter acquisition module is used to preprocess the three-dimensional point cloud data and input it into the pose fusion model to obtain the initial pose parameters.

[0050] The specific analysis method for preprocessing the three-dimensional point cloud data is: collect the three-dimensional point cloud data of the shield machine according to the sampling frequency set by the corresponding sensor and record the corresponding timestamp information, and adjust the three-dimensional point cloud data of the shield machine to the same time point.

[0051] It should be noted that the specific analysis method for adjusting to the same time point is: arranging the three-dimensional point cloud data of the shield machine according to the acquisition order, selecting the target time point , find two time points adjacent to the target time point from the arranged 3D point cloud data sequence, recorded as , obtain the corresponding original data value, and use the linear interpolation formula to calculate the estimated data value corresponding to the target time point , the linear interpolation formula is ,in Respectively represent the data values ​​corresponding to two adjacent time points of the target time point.

[0052] The specific analysis method of the posture parameter acquisition module is: according to the sampling frequency set by the corresponding sensor, the three-dimensional point cloud data of the shield machine are collected respectively and the corresponding timestamp information is recorded, and the three-dimensional point cloud data of the shield machine are adjusted to the same time point using a linear interpolation algorithm; it can smooth the data to a certain extent, fill in the missing or discontinuous information caused by time differences, improve the overall quality of the data, and facilitate more accurate analysis of the posture parameters of the shield machine, providing more reliable data support for the precise control of shield construction.

[0053] The Kalman filter model is selected as the posture fusion model, and the three-dimensional point cloud data of the shield machine is input to obtain the state estimation value of the shield machine. The iteration termination condition of the Kalman filter model is set as the change of the posture parameters is less than the preset threshold. According to the three-dimensional point cloud data of the shield machine and each state estimation value, the iterative calculation is entered. After the iterative calculation is completed, the three-dimensional position coordinates and attitude angle information of the shield machine are extracted from the state estimation value finally output by the model as the initial posture parameters; the posture of the shield machine is constantly changing during the construction process, and the iterative calculation method can adapt to this dynamic change well. By continuously updating data and iterative calculation, the model can capture the changes in the posture of the shield machine in time and accurately reflect its current true state.

[0054] It should be noted that the iterative calculation process of the Kalman filter includes two main steps: a prediction step and an update step. The three-dimensional point cloud data of the shield machine is converted into a linear form. In the prediction step, the control parameters and time interval of the shield machine are combined to predict the state estimate value at the next moment. The state estimate value at the next moment is predicted, and the covariance matrix at the next moment is predicted at the same time. In the update step, the Kalman gain is calculated, and the state estimate value is updated according to the observed value, and then the covariance matrix is ​​updated. After each iterative calculation is completed, check whether the change in the posture parameter is less than the preset threshold. If the termination condition is met, stop the iterative calculation, otherwise, continue to the next iteration. After the iterative calculation is completed, extract the three-dimensional position coordinates and posture angle of the shield machine from the final output state estimate value as the initial posture parameter.

[0055] The posture deviation analysis module is used to define the location of the shield machine in the three-dimensional geological model and select the relevant points of each axis for matching. The posture deviation parameters of the shield machine are evaluated by calculating the position deviation and posture deviation.

[0056] The specific analysis method for constructing the three-dimensional geological model is as follows: obtaining the three-dimensional geological point cloud data of the shield machine, clustering the point cloud data belonging to the same tunnel area into one category according to the regional division information of the surrounding tunnels, and extracting the point cloud data corresponding to each cluster; the geological conditions in the same tunnel area are often similar, and clustering them into one category can more accurately reflect the geological characteristics of each tunnel area, providing a more reliable data basis for the subsequent construction of the geological model.

[0057] The geometric surfaces are fitted according to the point cloud data corresponding to each cluster and combined to obtain a complete three-dimensional geological model. The geometric surfaces corresponding to each cluster are combined into a complete model to achieve the integration of geological data from different regions, so that the geological conditions of the entire shield construction area can be comprehensively analyzed, taking into account the mutual influence between different regions.

[0058] It should be noted that, in a specific embodiment, after the previous hierarchical clustering step, three types of point cloud data are obtained, corresponding to different geological areas: the first type of point cloud data is mainly concentrated in the top area of ​​the tunnel, showing a relatively flat distribution feature. After analysis, a plane is selected for fitting: the least squares method is used to calculate these point cloud data, and the equation of the plane is obtained as follows: ,in , that is, the plane equation is , indicating that the top of the tunnel is at a height of 10 meters above sea level.

[0059] The second type of point cloud data is distributed around the side wall of the tunnel and has certain cylindrical surface characteristics. The cylindrical surface fitting method is used to establish a mathematical model of the cylindrical surface and solve it using the least squares method to obtain the axis equation of the cylinder. (i.e. along the Z axis), with a radius of 3 meters, then the equation of the cylinder can be expressed as , describing the shape of the tunnel sidewalls.

[0060] The third type of point cloud data is located at the bottom area of ​​the tunnel, which also shows a relatively flat distribution. We select plane fitting and calculate the plane equation to be: , indicating that the bottom of the tunnel is at a height of 5 meters above sea level.

[0061] When these three fitted geometric surfaces (top plane, side wall cylindrical surface, and bottom plane) are combined, the top plane and the side wall cylindrical surface are smoothly connected at the top boundary, and the side wall cylindrical surface and the bottom plane can also smoothly transition at the bottom boundary, and finally a complete three-dimensional geological model of the tunnel area is obtained.

[0062] See also Figure 2 As shown, the specific analysis method of the relevant points of each axis is: obtain the front end point coordinates, the rear end point coordinates and the radius range of the shield machine from the real-time three-dimensional coordinates of the shield machine, take the axis of the shield machine as the center, and construct a cylinder in combination with the radius of the shield machine to define the range of the shield machine in the three-dimensional geological model; this provides a basis for the subsequent analysis of the interaction between the shield machine and the surrounding geological environment, and helps to timely discover potential geological risks.

[0063] Along the tunneling direction of the shield machine, the cross-sections of the shield machine within the range are divided according to the principle of equal spacing to obtain various cross-sections, and each of the cross-sections is perpendicular to the tunneling direction of the shield machine; the contact between the shield machine and the surrounding geology at different depths and positions can be studied in more detail, providing more detailed data for analyzing the stress and tunneling stability of the shield machine.

[0064] For each cross section, points located in the cross section are extracted from the three-dimensional geological point cloud data, and their distances to the center of the cross section are calculated respectively, wherein the center of the cross section is the projection point of the shield machine design axis on the cross section.

[0065] It should be noted that the center of the cross section is the projection point of the shield machine design axis on the cross section, and its coordinates are , for each point in the cross section , using the Euclidean distance formula Calculate its distance to the center of the cross section .

[0066] A reasonable deviation value is set, and the distance threshold is determined by summing it with the radius of the shield machine. The distance from each point to the center of the cross section is compared with the distance threshold, and the points whose distance to the center of the cross section is less than or equal to the distance threshold are screened out as the relevant points of each axis. By calculating the distance from the point to the center of the cross section and comparing it with the distance threshold, points closely related to the shield machine axis can be screened out. These points reflect the actual geological conditions around the shield machine and exclude some irrelevant points far away from the shield machine, so that subsequent analysis can focus more on key data.

[0067] It should be noted that the reasonable deviation value is usually set between 5% and 10% of the radius of the shield machine.

[0068] Linear fitting is performed on the axis-related points until the sum of the squares of the distances from the axis-related points to the fitting straight line is minimized, so as to determine the preliminary direction of the shield machine's excavation axis; the process of minimizing the sum of the squares of the distances will comprehensively consider the deviations of each point, so that the fitting straight line is as close as possible to the actual excavation axis trend, thereby obtaining a more accurate and reliable preliminary direction of the axis.

[0069] It should be noted that, assuming that the relevant points of each axis are Three-dimensional space points, whose coordinates are ,in Indicates The number of the relevant points of the axis, , let the straight line pass through the point , the direction vector is , then any point on the straight line It can be expressed as ,in , for a point in space and straight line ,point The distance to the line can be calculated by vector method, let vector , then point The square of the distance to the line is: ,for points, the sum of squared distances for: , take the centroid of all axis-related points as the initial point of the straight line , calculate the deviation vector of each point relative to the center of mass , construct the covariance matrix , then, the covariance matrix Perform eigenvalue decomposition and obtain the eigenvalue and the corresponding eigenvector , according to the least squares principle, the maximum eigenvalue The corresponding eigenvector This is the initial direction of the shield machine's excavation axis. .

[0070] The specific analysis method of the posture deviation parameters is as follows: randomly select two points on the excavation axis, obtain the direction vector of the excavation axis by calculating the vector difference formed by the two points, extract the axis-related points on the shield machine excavation axis, and simultaneously obtain the points on the preset design axis, find the nearest point on the design axis for each axis-related point, establish a corresponding relationship between the two, and form matching point pairs; the method of determining the direction vector with two points is simple and direct, avoiding complex direction calculations for a large number of axis-related points, and establishing matching point pairs at the same time so that each point in the actual excavation process can be accurately associated with the corresponding position under the design state, thereby providing an accurate data correspondence for the subsequent calculation of the position deviation.

[0071] It should be noted that, in a specific embodiment, 10 axis-related points are obtained, and the coordinates are , there are 20 points on the design axis , for a point on the excavation axis , find the closest point on the design axis through a multidimensional space search algorithm , then a matching point pair is formed , and corresponding matching points are found for the 10 axis-related points to form 10 matching point pairs.

[0072] The position difference of each matching point pair in three-dimensional space is calculated respectively, and the position deviation is obtained by taking the average value. The attitude deviation is obtained by calculating the angle between the design axis and the direction vector of the shield machine's excavation axis. The attitude deviation is obtained by calculating the angle between the design axis and the direction vector of the shield machine's excavation axis, which can clearly show the difference between the shield machine's attitude and the design attitude.

[0073] It should be noted that the posture deviation uses the vector dot product formula to calculate the dot product of the two direction vectors divided by the product of the module lengths of the two vectors to obtain the cosine value of the angle, and finally the actual angle between the two axes is calculated through inverse trigonometric functions. This angle is the posture deviation.

[0074] The posture deviation parameters of the shield machine are obtained based on the position deviation and posture deviation; the posture deviation parameters provide a clear goal and direction for the real-time adjustment of the shield machine, and the tunneling parameters of the shield machine can be adjusted in a targeted manner.

[0075] It should be noted that the specific calculation method of the posture deviation parameters of the shield machine is as follows: extract the position deviation and posture deviation of the shield machine, and then sum them up according to the weights to obtain the posture deviation parameters of the shield machine.

[0076] For example, the weights corresponding to the position deviation and attitude deviation of the shield machine are: .

[0077] The corrected posture parameter acquisition module is used to iteratively optimize the initial posture parameters and posture deviation parameters to obtain the corrected posture parameters.

[0078] The specific analysis method of the modified posture parameter acquisition module is: setting the state transfer matrix and the measurement matrix, taking the initial posture parameters of the shield machine as the initial value of the state vector, and predicting the state vector at the next moment by performing matrix multiplication operation on the initial value of the state vector and the state transfer matrix.

[0079] It should be noted that the state vector of the shield machine contains 6 elements, namely the three-dimensional position and 3D pose , under the assumption of simple uniform linear motion and unchanged posture, the state transfer matrix , attitude information measurement matrix , by performing matrix multiplication on the initial value of the state vector and the state transfer matrix, the state vector at the next moment can be predicted.

[0080] The posture deviation parameters are used as measurement values, and the predicted state vector is updated through the measurement matrix. Using the posture deviation parameters to update the state vector in real time can enable the system to track these changes in a timely manner, ensure that the state vector always reflects the latest status of the shield machine, and enhance the dynamic adaptability of the system.

[0081] The number of loop executions is determined according to the accuracy requirements of the shield machine's posture parameters. After the set number of iterations, the state vector obtained is recorded as the corrected posture parameter of the shield machine; multiple iterations can allow the state vector to gradually stabilize during the continuous updating process. Through repeated adjustments, the impact of accidental factors on state estimation is reduced, making the corrected posture parameters more reliable.

[0082] The posture instruction generation module is used to determine whether the corrected posture parameters exceed the dynamic threshold range. If so, a posture adjustment instruction is generated. Otherwise, trend analysis is performed on it together with historical data to output real-time posture warning information.

[0083] The specific analysis method of the posture instruction generation module is: setting a dynamic threshold range, comparing and judging the obtained corrected posture parameters with the dynamic threshold range, if the corrected posture parameters exceed the dynamic threshold range, generating a posture adjustment instruction, the posture adjustment instruction takes the direction exceeding the range as the positive direction, the adjustment direction is adjusted in the negative direction, the adjustment amplitude is determined according to the exceeding range, and the posture adjustment instruction is sent to the system to execute the corresponding posture adjustment operation; the adjustment direction is clearly defined as the opposite direction of the exceeding range, and the adjustment amplitude is determined according to the exceeding range, so that the adjustment operation is targeted and effective, and the shield machine posture deviation is avoided from further expansion.

[0084] If the corrected posture parameter does not exceed the dynamic threshold range, the corrected posture parameter is subjected to trend analysis with historical data, where the historical data is the posture angle data accumulated during the previous operation of the shield machine; this helps to gain a deeper understanding of the posture change characteristics of the shield machine at different construction stages.

[0085] The posture instruction generation module also includes: selecting a set number of time points as the time window size, organizing the historical data into a historical data sequence in order of time points, starting from the first data point of the historical data sequence, and taking the time window as the unit, calculating the average value of the posture angle data in each time window to obtain a moving average value sequence of the historical data sequence; the average value in each time window represents the overall level of the shield machine's posture angle in the time period, and the moving average value sequence reflects the stage-by-stage changes in the shield machine's posture angle over time, which is helpful to analyze the operating status of the shield machine in different time periods.

[0086] A linear regression is performed on the moving average sequence to calculate the slope of the regression line. If the slope of the regression line is positive, it indicates that the attitude angle of the shield machine is generally on an upward trend. If the slope of the regression line is negative, it indicates that the attitude angle of the shield machine is generally on a downward trend. If the slope of the regression line is within the set slope threshold range, it indicates that the stability of the attitude angle of the shield machine is acceptable, and the change of the attitude angle of the shield machine is output to the system as real-time posture warning information. When the slope is within the threshold range, it indicates that the change of the attitude angle of the shield machine is relatively stable and the stability is acceptable. Otherwise, it indicates that there may be abnormal changes in the attitude, which requires further attention and adjustment, which is helpful to timely discover potential problems in the operation of the shield machine and ensure construction safety and quality.

[0087] It should be noted that, in a specific embodiment, the shield machine has acquired the attitude angle data of 20 time points every 5 minutes in the past period of time, and the historical data sequence is: , set the number of time points to 4 as the time window size, starting from the first data point of the historical data sequence, divide the time window, the first time window is , whose average value is , the first time window is , the average value is , and so on, until the last time window, after calculation, the moving average sequence is: , perform linear regression analysis on the moving average sequence, and set the data point number in the moving average sequence as the independent variable , , corresponding to 15 moving average data points, the moving average is the dependent variable , the slope of the regression line obtained by linear regression is -0.07, and the slope threshold range is set to ,Since -0.07, that is, the slope of the regression line is negative and is not within the set slope threshold, it means that the attitude angle of the shield machine is generally on a downward trend, and the attitude angle change exceeds the stable range.

[0088] The management database is used to store the three-dimensional point cloud data, initial posture parameters, posture deviation parameters, corrected posture parameters and historical data of the shield machine body.

[0089] See also Figure 3 As shown, in addition, the present invention provides a shield machine posture measurement method, and the specific steps of the measurement method are as follows: S1. Shield machine data acquisition: Acquire three-dimensional point cloud data during the shield machine excavation process, and the three-dimensional point cloud data includes the real-time three-dimensional coordinates of the shield machine and three-dimensional geological point cloud data.

[0090] S2. Acquisition of pose parameters: Preprocess the 3D point cloud data and input it into the pose fusion model to obtain the initial pose parameters.

[0091] S3. Posture deviation analysis: define the location of the shield machine in the three-dimensional geological model and select the relevant points of each axis for matching, and evaluate the posture deviation parameters of the shield machine by calculating the position deviation and posture deviation.

[0092] S4. Obtaining corrected posture parameters: Iteratively optimize the initial posture parameters and posture deviation parameters to obtain corrected posture parameters.

[0093] S5. Posture instruction generation: Determine whether the corrected posture parameters exceed the dynamic threshold range. If so, generate a posture adjustment instruction. Otherwise, perform trend analysis on it together with historical data and output real-time posture warning information.

[0094] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations on the present invention. A person skilled in the art may make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention and they are still covered by the protection scope of the present invention.

Claims

1. A shield machine posture measurement system, characterized in that: The system specifically includes the following modules: The shield machine data acquisition module is used to obtain the 3D point cloud data of the shield machine during the tunneling process. The 3D point cloud data includes the real-time 3D coordinates of the shield machine and the 3D geological point cloud data; The pose parameter acquisition module is used to pre-process the 3D point cloud data and input it into the pose fusion model to obtain the initial pose parameters; The posture deviation analysis module is used to define the location of the shield machine in the three-dimensional geological model and select the relevant points of each axis for matching. The posture deviation parameters of the shield machine are evaluated by calculating the position deviation and posture deviation. A corrected posture parameter acquisition module is used to iteratively optimize the initial posture parameters and posture deviation parameters to obtain corrected posture parameters; The posture instruction generation module is used to determine whether the corrected posture parameters exceed the dynamic threshold range. If so, a posture adjustment instruction is generated. Otherwise, trend analysis is performed on it together with historical data to output real-time posture warning information. The management database is used to store the three-dimensional point cloud data, initial posture parameters, posture deviation parameters, corrected posture parameters and historical data of the shield machine body.

2. A shield machine posture measurement system according to claim 1, characterized in that: The specific analysis method for preprocessing the three-dimensional point cloud data is: According to the sampling frequency set by the corresponding sensor, the three-dimensional point cloud data of the shield machine are collected respectively and the corresponding timestamp information is recorded, and the three-dimensional point cloud data of the shield machine are adjusted to the same time point.

3. A shield machine posture measurement system according to claim 2, characterized in that: The specific analysis method of the posture parameter acquisition module is: The Kalman filter model is selected as the posture fusion model, and the three-dimensional point cloud data of the shield machine is input to obtain the state estimation value of the shield machine. The iteration termination condition of the Kalman filter model is set that the change of the posture parameter is less than the preset threshold. According to the three-dimensional point cloud data of the shield machine and each state estimation value, the iterative calculation is started. After the iterative calculation is completed, the three-dimensional position coordinates and attitude angle information of the shield machine are extracted from the state estimation value finally output by the model as the initial posture parameters.

4. A shield machine posture measurement system according to claim 2, characterized in that: The specific analysis method for constructing the three-dimensional geological model is: Obtain the three-dimensional geological point cloud data of the shield machine, cluster the point cloud data belonging to the same tunnel area into one category according to the regional division information of the surrounding tunnels, and extract the point cloud data corresponding to each cluster; Each geometric surface is fitted according to the point cloud data corresponding to each cluster, and they are combined to obtain a complete three-dimensional geological model.

5. A shield machine posture measurement system according to claim 4, characterized in that: The specific analysis method of the relevant points of each axis is as follows: The front end point coordinates, rear end point coordinates and radius range of the shield machine are obtained from the real-time three-dimensional coordinates of the shield machine. A cylinder is constructed with the axis of the shield machine as the center and the radius of the shield machine to define the range of the shield machine in the three-dimensional geological model. Along the tunneling direction of the shield machine, the shield machine in the range is divided into cross sections according to the principle of equal spacing to obtain each cross section, and each cross section is perpendicular to the tunneling direction of the shield machine; For each cross section, extract points located in the cross section from the three-dimensional geological point cloud data, and calculate the distances from the points to the center of the cross section, wherein the center of the cross section is the projection point of the shield machine design axis on the cross section; A reasonable deviation value is set, and the distance threshold is determined by summing it with the radius of the shield machine. The distance from each point to the center of the cross section is compared with the distance threshold, and the points whose distance to the center of the cross section is less than or equal to the distance threshold are selected as the relevant points of each axis. Linear fitting is performed on the axis-related points until the sum of the squares of the distances from the axis-related points to the fitting straight line reaches a minimum, thereby determining the preliminary direction of the shield machine's excavation axis.

6. A shield machine posture measurement system according to claim 5, characterized in that: The specific analysis method of the posture deviation parameters is: Take any two points on the tunneling axis, calculate the vector difference formed by the two points to obtain the direction vector of the tunneling axis, extract the axis-related points on the tunneling axis of the shield machine, and simultaneously obtain the points on the preset design axis, find the nearest point on the design axis for the axis-related points, establish the corresponding relationship between the two, and form matching point pairs; The position difference of each matching point pair in three-dimensional space is calculated respectively, and the position deviation is obtained by taking the average value. The attitude deviation is obtained by calculating the angle between the direction vector of the design axis and the shield machine excavation axis. The posture deviation parameters of the shield machine are evaluated based on position deviation and posture deviation.

7. A shield machine posture measurement system according to claim 6, characterized in that: The specific analysis method of the correction posture parameter acquisition module is: The state transfer matrix and the measurement matrix are set, and the initial posture parameters of the shield machine are used as the initial value of the state vector. The state vector at the next moment is predicted by performing matrix multiplication operation on the initial value of the state vector and the state transfer matrix. Taking the posture deviation parameter as the measurement value, the predicted state vector is updated through the measurement matrix; The number of loop executions is determined according to the accuracy requirements of the shield machine's posture parameters. After iterative operations with the set number of loop executions, the obtained state vector is recorded as the corrected posture parameter of the shield machine.

8. A shield machine posture measurement system according to claim 1, characterized in that: The specific analysis method of the posture instruction generation module is: A dynamic threshold range is set, and the obtained corrected posture parameters are compared with the dynamic threshold range. If the corrected posture parameters exceed the dynamic threshold range, a posture adjustment instruction is generated, wherein the posture adjustment instruction takes the direction exceeding the range as the positive direction, and the adjustment direction is adjusted in the negative direction. The adjustment amplitude is determined according to the exceeding range, and the posture adjustment instruction is sent to the system to execute the corresponding posture adjustment operation; If the corrected posture parameter does not exceed the dynamic threshold range, the corrected posture parameter is subjected to trend analysis with historical data, where the historical data is posture angle data accumulated during the previous operation of the shield machine.

9. A shield machine posture measurement system according to claim 8, characterized in that: The posture instruction generation module also includes: Select a set number of time points as the time window size, organize the historical data into a historical data sequence in the order of time points, and start from the first data point of the historical data sequence, taking the time window as a unit, calculate the average value of the posture angle data in each time window to obtain a moving average value sequence of the historical data sequence; Perform linear regression on the moving average sequence and calculate the slope of the regression line. If the slope of the regression line is positive, it means that the attitude angle of the shield machine is generally on an upward trend. If the slope of the regression line is negative, it means that the attitude angle of the shield machine is generally on a downward trend. If the slope of the regression line is within the set slope threshold range, it means that the stability of the attitude angle of the shield machine is qualified. The change of the attitude angle of the shield machine is output to the system as real-time posture warning information.

10. A shield machine posture measurement method, characterized in that: The steps include: S1. Shield machine data acquisition: Acquire the 3D point cloud data of the shield machine during excavation. The 3D point cloud data includes the real-time 3D coordinates of the shield machine and 3D geological point cloud data; S2. Acquisition of pose parameters: preprocessing the 3D point cloud data and inputting it into the pose fusion model to obtain the initial pose parameters; S3. Posture deviation analysis: define the location of the shield machine in the 3D geological model and select the relevant points of each axis for matching. Evaluate the posture deviation parameters of the shield machine by calculating the position deviation and posture deviation; S4. Acquisition of corrected posture parameters: iteratively optimize the initial posture parameters and posture deviation parameters to obtain corrected posture parameters; S5. Posture instruction generation: Determine whether the corrected posture parameters exceed the dynamic threshold range. If so, generate a posture adjustment instruction. Otherwise, perform trend analysis on it together with historical data and output real-time posture warning information.

Images