Visual monitoring method and system for cylinder construction attitude based on three-dimensional modeling

Abstract

The invention provides a cylinder construction attitude visual monitoring method and system based on three-dimensional modeling, and the method comprises the steps: collecting the three-dimensional point cloud data of a target cylinder, and constructing a digital twinborn model of the target cylinder based on the three-dimensional point cloud data; screening out surface key points; according to the surface key points, analyzing to obtain a current central axis of the target cylinder; calculating a current first offset vector based on the central axis and the design axis; calculating a current second offset vector according to the current coordinates and the design coordinates of the surface key points; determining an abnormal vector according to the vector length of the first offset vector and a preset length threshold; and according to the abnormal vector, the second offset vector and the position of the joint part, screening out joint key points, affected by dislocation, at the joint part from the surface key points, labeling the joint key points in a digital twinborn model, and visually displaying the labeled digital twinborn model. According to the invention, the real-time performance and efficiency of monitoring can be improved.

Description

Technical Field

[0001] This invention application relates to the field of 3D modeling, and in particular to a method and system for visual monitoring of the construction posture of a cylindrical structure based on 3D modeling. Background Technology

[0002] In the field of modern agricultural infrastructure construction, the construction quality and safety of grain silo structures directly affect the safety and economic benefits of grain storage, making it a crucial link in ensuring food security. Especially in the construction of large grain silos, the attitude control during silo construction is considered a key factor influencing the overall project stability and service life. However, current methods for monitoring the attitude of grain silo construction largely rely on manual measurement or simple instrument recording, generally lacking sufficient response to dynamic changes during construction. These methods often only acquire data at specific points in time, making it difficult to capture the continuity and real-time nature of attitude changes during construction. Particularly in continuous construction processes such as slipform construction, subtle changes in silo attitude can accumulate into significant deviations within a short period, thereby affecting structural safety.

[0003] Assessing the construction posture of the cylinder requires considering the offset of the central axis and the misalignment of the joints. The offset of the central axis directly affects the stress balance of various parts of the cylinder, leading to misalignment at some joints and creating structural safety hazards. Currently, existing technologies mainly rely on sensors to periodically collect changes in the central axis. After obtaining the data, further calculations and analyses of the joint changes are needed manually. This monitoring method suffers from low efficiency and a lack of intuitiveness. Summary of the Invention

[0004] This invention application provides a method and system for visual monitoring of the construction posture of a cylindrical structure based on 3D modeling, in order to solve the technical problem of how to improve the efficiency and intuitiveness of monitoring the construction posture of a cylindrical structure.

[0005] To address the aforementioned technical problems, this invention provides a method for visually monitoring the construction posture of a cylindrical structure based on 3D modeling, comprising: Collect three-dimensional point cloud data of the target cylinder, and construct a digital twin model of the target cylinder based on the three-dimensional point cloud data; Based on the digital twin model, key surface points of the target cylinder are selected; construction plan data of the target cylinder is obtained; wherein, the construction plan data includes preset cylinder model data, and the preset cylinder model data includes design axis and joint parts; Based on the surface key points, the current central axis of the target cylinder is obtained through analysis; and based on the selected surface key points, the design coordinates of each surface key point in the preset cylinder model are obtained through analysis of the preset cylinder model data. Based on the central axis and the design axis, the current first offset vector is calculated; based on the current coordinates and design coordinates of the surface key points, the current second offset vector of each surface key point is calculated respectively. When the first offset vector meets the preset conditions, based on the second offset vector and the position of the seam, the seam key points affected by the misalignment at the seam are selected from the surface key points; and each of the seam key points is marked on the digital twin model, the marked digital twin model is visualized, and the marked digital twin model is monitored.

[0006] As a preferred embodiment, the calculation of the current first offset vector based on the central axis and the design axis includes: The first offset vector is calculated by taking the intersection of the design axis and the preset cylindrical model as the starting point and the intersection of the central axis and the digital twin model as the ending point.

[0007] As a preferred embodiment, when the first offset vector satisfies a preset condition, the step of selecting the joint key points affected by misalignment at the joint location from the surface key points based on the second offset vector and the position of the joint location includes: Calculate the length of the first offset vector; When the length of the first offset vector is greater than a preset length threshold, the key points of the joint affected by the misalignment at the joint are selected from the key points of the surface based on the second offset vector and the position of the joint.

[0008] As a preferred embodiment, the step of selecting the joint key points affected by misalignment at the joint from the surface key points based on the second offset vector and the position of the joint includes: Based on the location of the seam, and combined with the distance between the surface key points and the seam, an initial set of seam points is selected from the surface key points. Calculate the radial and axial components of the second offset vector for each point in the initial joint point set; wherein the radial component is perpendicular to the design axis and the axial component is parallel to the design axis; Based on the distribution of the radial components, the radial misalignment region of the joint is determined by cluster analysis; wherein, the radial misalignment region is a continuous set of points where the radial component value exceeds a first preset threshold. Based on the distribution of the axial components, the axial misalignment region of the joint is determined by cluster analysis; wherein, the axial misalignment region is a continuous set of points where the axial component value exceeds a second preset threshold. The initial joint point located in the radial misalignment region is identified as the radial misalignment point, and the initial joint point in the axial misalignment region is identified as the axial misalignment point; and the radial misalignment point and the axial misalignment point are determined as the joint key points.

[0009] As a preferred embodiment, the construction plan data also includes construction error tolerance, and the preset cylinder model data also includes cylinder radius and cylinder height; the step of selecting an initial set of joint points from surface key points based on the location of the joint and the distance between the surface key points and the joint includes: Based on the preset cylindrical model data, the geometric features of the joint area are extracted, and the joint type is determined according to the geometric features; wherein, the geometric features include the joint length, and the joint type includes circumferential joints and longitudinal joints; When the joint type is a circumferential joint, the target screening range is determined based on the cylinder radius and construction error tolerance; when the joint type is a longitudinal joint, the target screening range is determined based on the ratio of joint length to cylinder height. In the digital twin model, a buffer area is constructed at the seam; wherein, the buffer area is a three-dimensional strip-shaped area centered at the seam and with the target screening range as its width; Key surface points located within the buffer area are selected to form an initial set of seam points.

[0010] As a preferred embodiment, the step of analyzing the current central axis of the target cylinder based on the key points on the surface includes: An initial reference coordinate system is established based on the direction of the design axis of the target cylinder. Each surface key point is projected onto the initial reference coordinate system, and the projection result of each surface key point is perpendicular to multiple equally spaced cross-sectional planes of the design axis, thus obtaining the projection point corresponding to each surface key point. Within each of the cross-sectional planes, the projection points are fitted with an ellipse using a random sampling consensus algorithm to obtain the fitting result; wherein, the number of samplings in the random sampling consensus algorithm is adaptively adjusted based on the distribution entropy value of the point cloud within the cross-sectional plane; Based on the fitting results, the spatial continuity between the elliptical centers of adjacent cross-sectional planes is calculated; and a smoothness constraint for the center point's motion trajectory is constructed based on the spatial continuity. Based on the smoothness constraints, the current central axis of the target cylinder is obtained according to the centers of each ellipse.

[0011] As a preferred embodiment, obtaining the current central axis of the target cylinder based on the centers of each of the ellipses includes: Using the centers of each ellipse as control points, a non-uniform rational B-spline curve is used for fitting to obtain the initial axis of the target cylinder; wherein, the parameters of the non-uniform rational B-spline curve are set according to the actual climbing trajectory of the slipform system during the construction of the target cylinder; The initial axis is optimized by iterative reweighted least squares method to obtain the current center axis of the target cylinder; wherein the weights of the iterative reweighted least squares method are dynamically adjusted according to the reciprocal of the average distance between the point cloud in each cross-sectional plane and the initial axis.

[0012] As a preferred embodiment, before determining the first offset vector as an anomaly vector, the method further includes: The length of the first offset vector is calculated using Euclidean distance.

[0013] As a preferred embodiment, constructing a digital twin model of the target cylinder based on the three-dimensional point cloud data includes: The three-dimensional point cloud data is segmented to obtain the cylindrical structural components of the target cylinder represented by the point cloud. Obtain the semantic information of the preset BIM model of the target cylinder; establish the association mapping between the cylinder structural components and the semantic information respectively; Based on the association mapping and the preset BIM model, a semantic digital model containing the topological connection relationships of components is constructed; The semantic digital model is updated by collecting real-time construction progress data to obtain the digital twin model.

[0014] Accordingly, this invention application also provides a visualization monitoring system for the construction posture of a cylindrical structure based on 3D modeling, including a model building module, a planned data acquisition module, a center axis analysis module, an offset vector calculation module, and a visualization monitoring module; wherein, The model building module is used to collect three-dimensional point cloud data of the target cylinder and build a digital twin model of the target cylinder based on the three-dimensional point cloud data. The planning data acquisition module is used to filter out key surface points of the target cylinder based on the digital twin model; and acquire construction plan data of the target cylinder; wherein, the construction plan data includes preset cylinder model data, and the preset cylinder model data includes design axis and joint parts; The central axis analysis module is used to analyze and obtain the current central axis of the target cylinder based on the surface key points; and to analyze and obtain the design coordinates of each surface key point in the preset cylinder model based on the preset cylinder model data based on the selected surface key points. The offset vector calculation module is used to calculate the current first offset vector based on the central axis and the design axis; and to calculate the current second offset vector of each surface key point according to the current coordinates and design coordinates of the surface key points. The visualization monitoring module is used to, when the first offset vector meets the preset conditions, select the joint key points affected by misalignment from the surface key points according to the second offset vector and the position of the joint; and annotate each of the joint key points in the digital twin model, visualize the annotated digital twin model, and monitor the annotated digital twin model.

[0015] As a preferred embodiment, the offset vector calculation module calculates the current first offset vector based on the central axis and the design axis, including: The offset vector calculation module takes the intersection of the design axis and the preset cylinder model as the starting point and the intersection of the central axis and the digital twin model as the ending point to calculate the first offset vector.

[0016] As a preferred embodiment, when the first offset vector meets a preset condition, the visualization monitoring module, based on the second offset vector and the position of the seam, filters out the seam key points affected by misalignment from the surface key points, including: The visualization monitoring module calculates the vector length of the first offset vector; When the length of the first offset vector is greater than a preset length threshold, the key points of the joint affected by the misalignment at the joint are selected from the key points of the surface based on the second offset vector and the position of the joint.

[0017] As a preferred embodiment, the visualization monitoring module, based on the second offset vector and the position of the seam, filters out the seam key points affected by misalignment from the surface key points, including: The visualization monitoring module selects an initial set of seam points from the surface key points based on the location of the seam and the distance between the surface key points and the seam. Calculate the radial and axial components of the second offset vector for each point in the initial joint point set; wherein the radial component is perpendicular to the design axis and the axial component is parallel to the design axis; Based on the distribution of the radial components, the radial misalignment region of the joint is determined by cluster analysis; wherein, the radial misalignment region is a continuous set of points where the radial component value exceeds a first preset threshold. Based on the distribution of the axial components, the axial misalignment region of the joint is determined by cluster analysis; wherein, the axial misalignment region is a continuous set of points where the axial component value exceeds a second preset threshold. The initial joint point located in the radial misalignment region is identified as the radial misalignment point, and the initial joint point in the axial misalignment region is identified as the axial misalignment point; and the radial misalignment point and the axial misalignment point are determined as the joint key points.

[0018] As a preferred embodiment, the construction plan data also includes construction error tolerance, and the preset cylinder model data also includes cylinder radius and cylinder height; the visualization monitoring module, based on the location of the joint and the distance between surface key points and the joint, filters out an initial set of joint points from the surface key points, including: The visualization monitoring module extracts the geometric features of the joints based on the preset cylinder model data, and determines the joint type according to the geometric features; wherein, the geometric features include the radius of curvature and the joint length, and the joint type includes circumferential joints and longitudinal joints; When the joint type is a circumferential joint, the target screening range is determined based on the cylinder radius and construction error tolerance; when the joint type is a longitudinal joint, the target screening range is determined based on the ratio of joint length to cylinder height. In the digital twin model, a buffer area is constructed at the seam; wherein, the buffer area is a three-dimensional strip-shaped area centered at the seam and with the target screening range as its width; Key surface points located within the buffer area are selected to form an initial set of seam points.

[0019] As a preferred embodiment, the central axis analysis module analyzes and obtains the current central axis of the target cylinder based on the surface key points, including: The central axis analysis module establishes an initial reference coordinate system based on the direction of the design axis of the target cylinder, projects each surface key point onto the initial reference coordinate system, and the projection results of each surface key point are perpendicular to multiple equally spaced cross-sectional planes of the design axis, thereby obtaining the projection points corresponding to each surface key point; Within each of the cross-sectional planes, the projection points are fitted with an ellipse using a random sampling consensus algorithm to obtain the fitting result; wherein, the number of samplings in the random sampling consensus algorithm is adaptively adjusted based on the distribution entropy value of the point cloud within the cross-sectional plane; Based on the fitting results, the spatial continuity between the elliptical centers of adjacent cross-sectional planes is calculated; and a smoothness constraint for the center point's motion trajectory is constructed based on the spatial continuity. Based on the smoothness constraints, the current central axis of the target cylinder is obtained according to the centers of each ellipse.

[0020] As a preferred embodiment, the central axis analysis module obtains the current central axis of the target cylinder based on the centers of each ellipse, including: The central axis analysis module uses the centers of each ellipse as control points and fits them with a non-uniform rational B-spline curve to obtain the initial axis of the target cylinder; wherein, the parameters of the non-uniform rational B-spline curve are set according to the actual climbing trajectory of the slipform system during the construction of the target cylinder; The initial axis is optimized by iterative reweighted least squares method to obtain the current center axis of the target cylinder; wherein the weights of the iterative reweighted least squares method are dynamically adjusted according to the reciprocal of the average distance between the point cloud in each cross-sectional plane and the initial axis.

[0021] As a preferred embodiment, the model building module constructs a digital twin model of the target cylinder based on the three-dimensional point cloud data, including: The model building module segments the three-dimensional point cloud data to obtain the cylindrical structural components of the target cylinder represented by the point cloud; Obtain the semantic information of the preset BIM model of the target cylinder; establish the association mapping between the cylinder structural components and the semantic information respectively; Based on the association mapping and the preset BIM model, a semantic digital model containing the topological connection relationships of components is constructed; The semantic digital model is updated by collecting real-time construction progress data to obtain the digital twin model.

[0022] Compared with the prior art, this invention application has the following beneficial effects: This invention application provides a method and system for visual monitoring of the construction posture of a cylindrical structure based on 3D modeling. The method includes: acquiring 3D point cloud data of the target cylindrical structure and constructing a digital twin model of the target cylindrical structure based on the 3D point cloud data; selecting key surface points of the target cylindrical structure based on the digital twin model; acquiring construction plan data of the target cylindrical structure; wherein the construction plan data includes preset cylindrical structure model data, which includes design axes and joint areas; analyzing the current center axis of the target cylindrical structure based on the key surface points; and based on the selected key surface points and the preset cylindrical structure model data... The analysis yields the design coordinates of each surface key point in the preset cylindrical model; based on the central axis and the design axis, the current first offset vector is calculated; according to the current coordinates and design coordinates of the surface key points, the current second offset vector of each surface key point is calculated; when the first offset vector meets the preset conditions, based on the second offset vector and the position of the joint, the joint key points affected by misalignment at the joint are selected from the surface key points; and each joint key point is marked in the digital twin model, the marked digital twin model is visualized, and the marked digital twin model is monitored. This invention application constructs a digital twin model of the target cylinder by collecting 3D point cloud data. The digital twin model reflects the current construction status of the cylinder and is compared with the construction plan data, enabling dynamic and visual monitoring of the construction status. This overcomes the problem of insufficient real-time performance of traditional methods that rely on periodic sensor data collection. Furthermore, by calculating the offset of the central axis (represented by a first offset vector) from the design axis in the preset cylinder model data, and calculating a second offset vector based on the current coordinates and design coordinates of key points on the surface, when the first offset vector meets preset conditions, the key points of the joint affected by misalignment are intelligently identified based on the second offset vector and the position of the joint. Compared with the existing technology that relies on manual calculation and judgment, this significantly improves monitoring efficiency. In addition, by visually annotating the key points of the joint in the digital twin model, relevant personnel can intuitively and quickly display abnormalities in the construction posture, improving the intuitiveness of the monitoring results through visualization. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating an embodiment of the method for visual monitoring of the construction posture of a cylindrical structure based on three-dimensional modeling provided in this invention application.

[0024] Figure 2 This is a flowchart illustrating a preferred embodiment of the method for visual monitoring of the construction posture of a cylindrical structure based on three-dimensional modeling provided in this invention application.

[0025] Figure 3This is a flowchart illustrating a preferred embodiment of the method for visual monitoring of the construction posture of a cylindrical structure based on three-dimensional modeling provided in this invention application.

[0026] Figure 4 This is a flowchart illustrating a preferred embodiment of the method for visual monitoring of the construction posture of a cylindrical structure based on 3D modeling provided in this invention application.

[0027] Figure 5 This is a flowchart illustrating a preferred embodiment of the method for visual monitoring of the construction posture of a cylindrical structure based on three-dimensional modeling provided in this invention application.

[0028] Figure 6 This is a structural schematic diagram of an embodiment of the cylindrical construction posture visualization monitoring system based on three-dimensional modeling provided in this invention application. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] Example 1 According to relevant technical records, ensuring construction quality and structural stability is a crucial core task in building construction, especially in high-precision projects such as large grain silo structures. These projects not only relate to the safety of grain storage but also directly affect the durability and economic benefits of subsequent use. However, with the expansion of project scale and increasing complexity, how to monitor the construction posture of the silo structure in real time during construction and promptly identify potential problems has become a key area that urgently needs to be addressed within the industry.

[0031] Currently, although many construction teams employ traditional surveying tools and techniques to monitor the construction status of the cylinder, these methods often have significant limitations. Especially when facing complex construction environments, relying solely on manual measurement or localized inspections makes it difficult to comprehensively capture subtle changes in the overall cylinder, let alone quickly compare these changes with specifications. This approach proves inadequate when dealing with large-scale, multi-dimensional construction information, often leading to delayed problem detection and even impacting project progress.

[0032] The assessment of the cylinder's construction posture requires precise control of several key parameters, with core considerations including the offset of the central axis and the misalignment of the segment joints. The offset of the central axis is a crucial indicator of the overall verticality and stability of the cylinder; any deviation directly affects the stress balance of different parts of the cylinder, potentially leading to structural safety hazards. This offset can further cause misalignment at the segment joints, reducing the strength of the cylinder at the connection points and potentially leading to cracks or deformation during long-term use. For example, in actual construction, if the central axis of a segment deviates from the design value by several centimeters, it may result in an inability to achieve a tight seal at the joints during subsequent segment installation, creating stress concentration points and ultimately posing a safety hazard.

[0033] Therefore, how to quantify the changes of these key parameters in real time during construction and present the complex measurement results to construction personnel in an intuitive way has become a key issue in improving the accuracy of project quality control. Solving this problem requires not only overcoming technical barriers but also finding effective application methods in actual business to cope with the multiple challenges in complex construction scenarios.

[0034] Regarding one or more of the above technical issues, such as Figure 1 As shown, Figure 1 This is a flowchart illustrating an embodiment of the method for visual monitoring of the construction posture of a cylindrical structure based on three-dimensional modeling provided in this invention application.

[0035] In this embodiment, the method for visual monitoring of the construction posture of the cylinder based on three-dimensional modeling can be applied to computer equipment, including but not limited to smartphones, laptops, tablets, desktop computers, as well as physical servers and cloud servers connected to display units.

[0036] After an image acquisition device, such as a camera, acquires an image of the target cylinder, or after a laser scanner obtains point cloud data of the target cylinder, the aforementioned computer device can execute the 3D modeling-based cylinder construction posture visualization monitoring method and control the display unit to visualize the monitoring results.

[0037] Figure 1 The illustrated embodiment includes steps S101 to S105; each step is described in detail below: Step S101: Collect three-dimensional point cloud data of the target cylinder, and construct a digital twin model of the target cylinder based on the three-dimensional point cloud data.

[0038] In this step, three-dimensional point cloud data of the target cylinder can be collected by using laser scanners, lidar, or sensor networks pre-deployed at the construction site.

[0039] In some examples, the aforementioned sensor network may include high-definition cameras to acquire images of the target cylinder, thereby assisting in the construction of a digital twin model using 3D point cloud data.

[0040] For example, the aforementioned lidar can collect 1,000 point cloud data points per second with an accuracy of 0.01 meters, the camera resolution is 1920x1080 pixels, the acquisition frequency is 30 frames per second, and the data is transmitted to the cloud server through a 5G network to ensure a latency of less than 50 milliseconds.

[0041] Image processing algorithms were used to perform preliminary filtering and feature extraction on the acquired data. Specifically, the Gaussian blur algorithm in the OpenCV library was used to remove image noise, with the Gaussian kernel size set to 5x5 and the standard deviation set to 1.5. Then, the Canny edge detection algorithm was used to extract the edge features of the cylinder surface, with the low threshold set to 50 and the high threshold set to 150, resulting in clear edge contours.

[0042] Using the Harris corner detection algorithm with a corner response threshold of 0.1, the coordinates of key points on the cylinder surface were detected. Assuming approximately 200 key points are extracted per frame with a coordinate accuracy of 0.1 pixels, random errors are reduced by averaging over 10 consecutive frames, resulting in a stable set of key point coordinates. These key point coordinates are then fused and matched with LiDAR point cloud data to construct a digital twin model.

[0043] In other preferred embodiments, such as Figure 2 As shown, step S101, which involves constructing a digital twin model of the target cylinder based on the three-dimensional point cloud data, includes steps S201 to S204; each step is detailed below: Step S201: Segment the three-dimensional point cloud data to obtain the cylindrical structure components of the target cylinder represented by the point cloud; Step S202: Obtain the semantic information of the preset BIM model of the target cylinder; establish the association mapping between the cylinder structural components and the semantic information respectively; Step S203: Based on the association mapping and the preset BIM model, construct a semantic digital model containing the topological connection relationships of components; Step S204: Update the semantic digital model using real-time collected construction progress data to obtain the digital twin model.

[0044] In this embodiment, BIM model should be referred to as Building Information Modeling in English. This preferred embodiment refines the construction process of the digital twin model. By segmenting 3D point cloud data to obtain the structural components of the target tube and associating them with the semantic information of the preset BIM model, a semantic digital model containing topological connections is constructed. The digital twin model is obtained by updating the data with real-time construction progress data. Compared with the existing technology where the digital model is only built based on point cloud, this can solve the problem of the lack of semantic association and dynamic update capability in the existing technology. This technical feature makes the digital twin model more practical. In addition, the semantic association and topological connection of BIM ensure that the model can accurately reflect the logical relationship between components. The real-time construction progress data update keeps the model synchronized with the actual construction status, providing a reliable model foundation for the accurate selection of surface key points and the comparison of construction plan data. It avoids monitoring deviations caused by the disconnect between the model and reality, and can effectively improve the effectiveness of dynamic monitoring.

[0045] Step S102: Based on the digital twin model, select the key points on the surface of the target cylinder; obtain the construction plan data of the target cylinder.

[0046] The construction plan data includes preset cylinder model data, which includes design axis and joint locations.

[0047] It should be noted that the difference between the preset cylinder model and the digital twin model is that the digital twin model reflects the actual construction progress and dimensions of the target cylinder, while the preset cylinder model is generated according to the construction plan and reflects the standard model of the target cylinder.

[0048] The design axis is the central axis of the pre-designed cylindrical model, while the joint area is the location where joints are to be made according to the needs of the construction plan.

[0049] In this step, based on the digital twin model, data such as the shape and size of part of the target cylinder can be obtained, thereby determining several key surface points.

[0050] For example, the selection of surface key points can prioritize shape inflection points, corner points, or contour points, and then consider points on planes or near-planes.

[0051] Step S103: Based on the surface key points, analyze and obtain the current central axis of the target cylinder; and based on the selected surface key points, analyze the design coordinates of each surface key point in the preset cylinder model based on the preset cylinder model data.

[0052] In some embodiments, the current central axis of the target cylinder can be determined by analyzing the positional patterns of key points on the surface. As the visualization monitoring time progresses, the position of the central axis may change or shift. By continuously monitoring the actual position of the central axis, the degree of change in the construction posture can be determined.

[0053] Furthermore, considering that the surface key points are generated based on the digital twin model according to the actual construction situation, this step can determine the design coordinates of these surface key points in the preset cylinder model based on the preset mapping relationship when determining the surface key points.

[0054] In some preferred solutions, such as Figure 3 As shown, the step of analyzing the current central axis of the target cylinder based on the key points on the surface includes steps S301 to S304; each step is detailed below: Step S301: Establish an initial reference coordinate system based on the direction of the design axis of the target cylinder, project each surface key point onto the initial reference coordinate system, and the projection result of each surface key point is perpendicular to multiple equally spaced cross-sectional planes of the design axis, thereby obtaining the projection point corresponding to each surface key point.

[0055] Step S302: Within each of the cross-sectional planes, the projection points are fitted with an ellipse using a random sampling consensus algorithm to obtain a fitting result; wherein, the number of samplings in the random sampling consensus algorithm is adaptively adjusted based on the distribution entropy value of the point cloud within the cross-sectional plane.

[0056] Step S303: Based on the fitting results, calculate the spatial continuity between the elliptical centers of adjacent cross-sectional planes; and construct a smoothness constraint for the center point's motion trajectory based on the spatial continuity.

[0057] Step S304: Based on the smoothness constraint, the current central axis of the target cylinder is obtained according to the centers of each ellipse.

[0058] This preferred embodiment refines the method for analyzing the current central axis of the target cylinder. Compared with the existing technology, which suffers from fixed sampling parameters and neglect of spatial continuity in axis calculation, this embodiment establishes an initial reference coordinate system, projects key points on the surface onto the cross-sectional plane, and uses a random sampling consensus algorithm that adaptively adjusts the number of samplings based on the point cloud distribution entropy value for ellipse fitting. Combined with the spatial continuity of the ellipse center to construct a smoothness constraint, the calculation of the central axis can be made more accurate and stable. In addition, the adaptive sampling number can adapt to the point cloud distribution state of different cross-sections, avoid interference from abnormal points, and the spatial continuity and smoothness constraints ensure the rationality of the axis, providing a reliable benchmark for the calculation of the first offset vector, reducing monitoring misjudgments caused by axis deviation, and improving the accuracy of construction posture monitoring.

[0059] Further, obtaining the current central axis of the target cylinder based on the centers of each ellipse includes: using each ellipse center as a control point, fitting a non-uniform rational B-spline curve to obtain the initial axis of the target cylinder; wherein the parameters of the non-uniform rational B-spline curve are set according to the actual climbing trajectory of the slipform system during the construction of the target cylinder; optimizing the initial axis using an iterative reweighted least squares method to obtain the current central axis of the target cylinder; wherein the weights of the iterative reweighted least squares method are dynamically adjusted based on the reciprocal of the average distance between the point cloud in each cross-sectional plane and the initial axis.

[0060] Compared to existing axis optimization methods with fixed fitting methods and single weights, this preferred embodiment optimizes the final generation method of the center axis. Specifically, it obtains the initial axis by fitting a non-uniform rational B-spline curve with the actual climbing trajectory parameters of the sliding formwork system, and then optimizes it using an iterative reweighted least squares method with dynamically adjusted weights, making the center axis more closely match the actual construction. In addition, the combination of B-spline curve parameters and sliding formwork trajectory ensures that the axis conforms to the construction process logic. The dynamically adjusted optimization weights can highlight the role of effective point clouds, suppress the influence of outliers, and correct minor deviations in the initial axis, making the obtained center axis more accurately reflect the actual construction posture of the cylinder, and providing a more accurate current state benchmark for the calculation of the first offset vector.

[0061] Step S104: Calculate the current first offset vector based on the central axis and the design axis; calculate the current second offset vector of each surface key point according to the current coordinates and design coordinates of the surface key points.

[0062] In some preferred embodiments, when calculating the second offset vector, the design coordinates can be used as the starting point and the current coordinates as the ending point to calculate the second offset vector.

[0063] When calculating the first offset vector, considering that the central axis and the design axis are often a straight line, in some embodiments the intersection of the design axis and the preset cylindrical model can be used as the starting point, and the intersection of the central axis and the digital twin model can be used as the ending point to calculate the first offset vector.

[0064] In other embodiments, the intersection of the design axis and the two ends of the preset cylindrical model can be determined as the first line segment, and the intersection of the central axis and the two ends of the digital twin model can be determined as the second line segment. The midpoint of the first line segment is taken as the starting point, and the midpoint of the second line segment is taken as the ending point, and the aforementioned first offset vector is calculated.

[0065] Step S105: When the first offset vector meets the preset conditions, based on the second offset vector and the position of the seam, select the seam key points affected by misalignment from the surface key points; and mark each of the seam key points in the digital twin model, visualize the marked digital twin model, and monitor the marked digital twin model.

[0066] In this step, the aforementioned preset conditions can be determined based on the vector length of the first offset vector.

[0067] The length of the aforementioned vector can also be called the magnitude or modulus of the vector, which mathematically refers to the displacement of the vector in its direction. In this embodiment, it can be calculated using Euclidean distance.

[0068] Further, step S105, which involves selecting the joint key points affected by misalignment at the joint from the surface key points based on the second offset vector and the position of the joint when the first offset vector meets the preset conditions, includes: calculating the vector length of the first offset vector; and when the vector length of the first offset vector is greater than a preset length threshold, selecting the joint key points affected by misalignment at the joint from the surface key points based on the second offset vector and the position of the joint.

[0069] In some preferred embodiments, such as Figure 4 As shown, step S105, which involves selecting the joint key points affected by misalignment from the surface key points based on the second offset vector and the position of the joint, includes steps S401 to S405; each step is detailed below: Step S401: Based on the location of the seam and the distance between the surface key points and the seam, select an initial seam point set from the surface key points. Step S402: Calculate the radial and axial components of the second offset vector for each point in the initial joint point set; wherein the radial component is perpendicular to the design axis and the axial component is parallel to the design axis. Step S403: Based on the distribution of the radial components, determine the radial misalignment region of the joint through cluster analysis; wherein, the radial misalignment region is a continuous set of points where the radial component value exceeds a first preset threshold. Step S404: Based on the distribution of the axial components, determine the axial misalignment region of the joint through cluster analysis; wherein, the axial misalignment region is a continuous set of points where the axial component value exceeds a second preset threshold. Step S405: Identify the initial joint point located in the radial misalignment area as the radial misalignment point, and identify the initial joint point in the axial misalignment area as the axial misalignment point; and determine the radial misalignment point and the axial misalignment point as the joint key points.

[0070] This preferred embodiment decomposes the second offset vector into radial and axial components and performs directional analysis based on the design axis of the cylindrical structure. This avoids the limitations of conventional methods that rely solely on vector magnitude and can distinguish different types of misalignment effects (such as radial expansion or axial displacement), thus reflecting construction deviations at the joints more precisely. Furthermore, cluster analysis identifies radial and axial misalignment areas at the joints, making the visualization of subsequent steps more detailed and structured. Relevant personnel can intuitively see the type and location of misalignment, for example, by marking radial and axial misalignments with different colors in the digital twin model, thereby quickly understanding the root cause of the problem and taking targeted measures, improving the practicality and operability of the monitoring results. Moreover, by processing the distribution of components through cluster analysis, continuous misalignment areas are identified, rather than isolated points. This is more consistent with the diffusion characteristics of misalignment in actual construction, improving the accuracy and intelligence of the screening.

[0071] Furthermore, when determining the initial set of joint points, the joint type and the geometric features of the joint location can be combined to ensure that the selected initial set of joint points matches the characteristics of the cylinder structure.

[0072] For example, such as Figure 5 As shown, Figure 5 The construction plan data of the illustrated embodiment also includes construction error tolerance, and the preset cylinder model data also includes cylinder radius and cylinder height.

[0073] Step S401 involves selecting an initial set of seam points from the surface key points based on the location of the seam and the distance between the surface key points and the seam, including steps S501 to S504; each step is detailed below: Step S501: Based on the preset cylinder model data, extract the geometric features of the joint area, and determine the joint type according to the geometric features; wherein, the geometric features include the radius of curvature and the joint length, and the joint type includes circumferential joints and longitudinal joints; Step S502: When the joint type is a circumferential joint, the target screening range is determined based on the cylinder radius and construction error tolerance; when the joint type is a longitudinal joint, the target screening range is determined based on the ratio of joint length to cylinder height. Step S503: In the digital twin model, a buffer area for the seam is constructed; wherein, the buffer area is a three-dimensional strip-shaped area centered on the seam and with the target screening range as its width; Step S504: Select the key surface points located within the buffer area to form an initial set of seam points.

[0074] Compared with the existing technical solution of filtering point sets by a fixed distance threshold, this preferred embodiment determines the target filtering range based on the joint type (circumferential / longitudinal) and the cylinder radius and cylinder height, so that the distance filtering matches the cylinder structural characteristics; in addition, through adaptive analysis of joint features, the distance-based filtering can be more consistent with the actual cylinder structure.

[0075] Furthermore, in step S504, the local density of each point in the point set can be calculated. If the local density is lower than a preset density threshold, the target screening range is adaptively adjusted and the points are re-screened until the distribution of the point set meets the requirements of uniformity and continuity.

[0076] Accordingly, such as Figure 6 As shown, this invention application also provides a 3D modeling-based visualization monitoring system 600 for the construction posture of a cylindrical structure, including a model building module 601, a planned data acquisition module 602, a center axis analysis module 603, an offset vector calculation module 604, and a visualization monitoring module 605; wherein, The model building module 601 is used to collect three-dimensional point cloud data of the target cylinder and build a digital twin model of the target cylinder based on the three-dimensional point cloud data. The planning data acquisition module 602 is used to filter out key surface points of the target cylinder based on the digital twin model; and acquire construction planning data of the target cylinder; wherein, the construction planning data includes preset cylinder model data, and the preset cylinder model data includes design axis and joint parts; The central axis analysis module 603 is used to analyze and obtain the current central axis of the target cylinder based on the surface key points; and to analyze and obtain the design coordinates of each surface key point in the preset cylinder model based on the preset cylinder model data based on the selected surface key points. The offset vector calculation module 604 is used to calculate the current first offset vector based on the central axis and the design axis; and to calculate the current second offset vector of each surface key point according to the current coordinates and design coordinates of the surface key points. The visualization monitoring module 605 is used to, when the first offset vector meets the preset conditions, select the joint key points affected by misalignment from the surface key points according to the second offset vector and the position of the joint; and mark each of the joint key points in the digital twin model, visualize the marked digital twin model, and monitor the marked digital twin model.

[0077] As a preferred embodiment, the offset vector calculation module 604 calculates the current first offset vector based on the central axis and the design axis, including: The offset vector calculation module 604 takes the intersection of the design axis and the preset cylinder model as the starting point and the intersection of the central axis and the digital twin model as the ending point to calculate the first offset vector.

[0078] As a preferred embodiment, when the first offset vector meets the preset conditions, the visualization monitoring module 605, based on the second offset vector and the position of the joint, filters out the joint key points affected by misalignment from the surface key points, including: The visualization monitoring module 605 calculates the vector length of the first offset vector; When the length of the first offset vector is greater than a preset length threshold, the key points of the joint affected by the misalignment at the joint are selected from the key points of the surface based on the second offset vector and the position of the joint.

[0079] As a preferred embodiment, the visualization monitoring module 605, based on the second offset vector and the position of the seam, filters out the seam key points affected by misalignment from the surface key points, including: The visualization monitoring module 605 selects an initial set of seam points from the surface key points based on the location of the seam and the distance between the surface key points and the seam. Calculate the radial and axial components of the second offset vector for each point in the initial joint point set; wherein the radial component is perpendicular to the design axis and the axial component is parallel to the design axis; Based on the distribution of the radial components, the radial misalignment region of the joint is determined by cluster analysis; wherein, the radial misalignment region is a continuous set of points where the radial component value exceeds a first preset threshold. Based on the distribution of the axial components, the axial misalignment region of the joint is determined by cluster analysis; wherein, the axial misalignment region is a continuous set of points where the axial component value exceeds a second preset threshold. The initial joint point located in the radial misalignment region is identified as the radial misalignment point, and the initial joint point in the axial misalignment region is identified as the axial misalignment point; and the radial misalignment point and the axial misalignment point are determined as the joint key points.

[0080] As a preferred embodiment, the construction plan data also includes construction error tolerance, and the preset cylinder model data also includes cylinder radius and cylinder height; the visualization monitoring module 605, based on the location of the joint and the distance between the surface key points and the joint, filters out an initial set of joint points from the surface key points, including: The visualization monitoring module 605 extracts the geometric features of the joint area based on the preset cylinder model data, and determines the joint type according to the geometric features; wherein, the geometric features include the radius of curvature and the joint length, and the joint type includes circumferential joints and longitudinal joints; When the joint type is a circumferential joint, the target screening range is determined based on the cylinder radius and construction error tolerance; when the joint type is a longitudinal joint, the target screening range is determined based on the ratio of joint length to cylinder height. In the digital twin model, a buffer area is constructed at the seam; wherein, the buffer area is a three-dimensional strip-shaped area centered at the seam and with the target screening range as its width; Key surface points located within the buffer area are selected to form an initial set of seam points.

[0081] As a preferred embodiment, the central axis analysis module 603 analyzes and obtains the current central axis of the target cylinder based on the surface key points, including: The central axis analysis module 603 establishes an initial reference coordinate system based on the direction of the design axis of the target cylinder, projects each surface key point onto the initial reference coordinate system, and the projection results of each surface key point are perpendicular to multiple equally spaced cross-sectional planes of the design axis, thereby obtaining the projection points corresponding to each surface key point. Within each of the cross-sectional planes, the projection points are fitted with an ellipse using a random sampling consensus algorithm to obtain the fitting result; wherein, the number of samplings in the random sampling consensus algorithm is adaptively adjusted based on the distribution entropy value of the point cloud within the cross-sectional plane; Based on the fitting results, the spatial continuity between the elliptical centers of adjacent cross-sectional planes is calculated; and a smoothness constraint for the center point's motion trajectory is constructed based on the spatial continuity. Based on the smoothness constraints, the current central axis of the target cylinder is obtained according to the centers of each ellipse.

[0082] As a preferred embodiment, the central axis analysis module 603 obtains the current central axis of the target cylinder based on the centers of each ellipse, including: The central axis analysis module 603 uses the centers of each ellipse as control points and fits them with a non-uniform rational B-spline curve to obtain the initial axis of the target cylinder; wherein, the parameters of the non-uniform rational B-spline curve are set according to the actual climbing trajectory of the slipform system during the construction of the target cylinder. The initial axis is optimized by iterative reweighted least squares method to obtain the current center axis of the target cylinder; wherein the weights of the iterative reweighted least squares method are dynamically adjusted according to the reciprocal of the average distance between the point cloud in each cross-sectional plane and the initial axis.

[0083] As a preferred embodiment, the model building module 601 constructs a digital twin model of the target cylinder based on the three-dimensional point cloud data, including: The model building module 601 segments the three-dimensional point cloud data to obtain the cylindrical structure components of the target cylinder represented by the point cloud; Obtain the semantic information of the preset BIM model of the target cylinder; establish the association mapping between the cylinder structural components and the semantic information respectively; Based on the association mapping and the preset BIM model, a semantic digital model containing the topological connection relationships of components is constructed; The semantic digital model is updated by collecting real-time construction progress data to obtain the digital twin model.

[0084] Compared with the prior art, this invention application has the following beneficial effects: This invention application provides a method and system for visual monitoring of the construction posture of a cylindrical structure based on 3D modeling. The method includes: acquiring 3D point cloud data of the target cylindrical structure and constructing a digital twin model of the target cylindrical structure based on the 3D point cloud data; selecting key surface points of the target cylindrical structure based on the digital twin model; acquiring construction plan data of the target cylindrical structure; wherein the construction plan data includes preset cylindrical structure model data, which includes design axes and joint areas; analyzing the current center axis of the target cylindrical structure based on the key surface points; and based on the selected key surface points and the preset cylindrical structure model data... The analysis yields the design coordinates of each surface key point in the preset cylindrical model; based on the central axis and the design axis, the current first offset vector is calculated; according to the current coordinates and design coordinates of the surface key points, the current second offset vector of each surface key point is calculated; when the first offset vector meets the preset conditions, based on the second offset vector and the position of the joint, the joint key points affected by misalignment at the joint are selected from the surface key points; and each joint key point is marked in the digital twin model, the marked digital twin model is visualized, and the marked digital twin model is monitored. This invention application constructs a digital twin model of the target cylinder by collecting 3D point cloud data. The digital twin model reflects the current construction status of the cylinder and is compared with the construction plan data, enabling dynamic and visual monitoring of the construction status. This overcomes the problem of insufficient real-time performance of traditional methods that rely on periodic sensor data collection. Furthermore, by calculating the offset of the central axis (represented by a first offset vector) from the design axis in the preset cylinder model data, and calculating a second offset vector based on the current coordinates and design coordinates of key points on the surface, when the first offset vector meets preset conditions, the key points of the joint affected by misalignment are intelligently identified based on the second offset vector and the position of the joint. Compared with the existing technology that relies on manual calculation and judgment, this significantly improves monitoring efficiency. In addition, by visually annotating the key points of the joint in the digital twin model, relevant personnel can intuitively and quickly display abnormalities in the construction posture, improving the intuitiveness of the monitoring results through visualization.

[0085] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for visually monitoring the construction posture of a cylindrical structure based on 3D modeling, characterized in that, include: Collect three-dimensional point cloud data of the target cylinder, and construct a digital twin model of the target cylinder based on the three-dimensional point cloud data; Based on the digital twin model, key surface points of the target cylinder are selected; construction plan data of the target cylinder is obtained; wherein, the construction plan data includes preset cylinder model data, and the preset cylinder model data includes design axis and joint parts; Based on the surface key points, the current central axis of the target cylinder is obtained through analysis; and based on the selected surface key points, the design coordinates of each surface key point in the preset cylinder model are obtained through analysis of the preset cylinder model data. Based on the central axis and the design axis, the current first offset vector is calculated; based on the current coordinates and design coordinates of the surface key points, the current second offset vector of each surface key point is calculated respectively. When the first offset vector meets the preset conditions, based on the second offset vector and the position of the seam, the seam key points affected by the misalignment at the seam are selected from the surface key points; and each of the seam key points is marked on the digital twin model, the marked digital twin model is visualized, and the marked digital twin model is monitored.

2. The method for visual monitoring of cylinder construction posture based on three-dimensional modeling as described in claim 1, characterized in that, The calculation of the current first offset vector based on the central axis and the design axis includes: The first offset vector is calculated by taking the intersection of the design axis and the preset cylindrical model as the starting point and the intersection of the central axis and the digital twin model as the ending point.

3. The method for visual monitoring of the construction posture of a cylindrical structure based on three-dimensional modeling as described in claim 2, characterized in that, When the first offset vector meets the preset condition, based on the second offset vector and the position of the seam, the key seam points affected by misalignment at the seam are selected from the surface key points, including: Calculate the length of the first offset vector; When the length of the first offset vector is greater than a preset length threshold, the key points of the joint affected by the misalignment at the joint are selected from the key points of the surface based on the second offset vector and the position of the joint.

4. The method for visual monitoring of cylinder construction posture based on three-dimensional modeling as described in claim 1, characterized in that, The step of selecting the joint key points affected by misalignment at the joint location from the surface key points based on the second offset vector and the position of the joint includes: Based on the location of the seam, and combined with the distance between the surface key points and the seam, an initial set of seam points is selected from the surface key points. Calculate the radial and axial components of the second offset vector for each point in the initial joint point set; wherein the radial component is perpendicular to the design axis and the axial component is parallel to the design axis; Based on the distribution of the radial components, the radial misalignment region of the joint is determined by cluster analysis; wherein, the radial misalignment region is a continuous set of points where the radial component value exceeds a first preset threshold. Based on the distribution of the axial components, the axial misalignment region of the joint is determined by cluster analysis; wherein, the axial misalignment region is a continuous set of points where the axial component value exceeds a second preset threshold. The initial joint point located in the radial misalignment region is identified as the radial misalignment point, and the initial joint point in the axial misalignment region is identified as the axial misalignment point; and the radial misalignment point and the axial misalignment point are determined as the joint key points.

5. The method for visual monitoring of the construction posture of a cylindrical structure based on three-dimensional modeling as described in claim 1, characterized in that, The step of analyzing the current central axis of the target cylinder based on the surface key points includes: An initial reference coordinate system is established based on the direction of the design axis of the target cylinder. Each surface key point is projected onto the initial reference coordinate system, and the projection result of each surface key point is perpendicular to multiple equally spaced cross-sectional planes of the design axis, thus obtaining the projection point corresponding to each surface key point. Within each of the cross-sectional planes, the projection points are fitted with an ellipse using a random sampling consensus algorithm to obtain the fitting result; wherein, the number of samplings in the random sampling consensus algorithm is adaptively adjusted based on the distribution entropy value of the point cloud within the cross-sectional plane; Based on the fitting results, the spatial continuity between the elliptical centers of adjacent cross-sectional planes is calculated; and a smoothness constraint for the center point's motion trajectory is constructed based on the spatial continuity. Based on the smoothness constraints, the current central axis of the target cylinder is obtained according to the centers of each ellipse.

6. The method for visual monitoring of cylinder construction posture based on three-dimensional modeling as described in claim 5, characterized in that, The step of obtaining the current central axis of the target cylinder based on the centers of each of the ellipses includes: Using the centers of each ellipse as control points, a non-uniform rational B-spline curve is used for fitting to obtain the initial axis of the target cylinder; wherein, the parameters of the non-uniform rational B-spline curve are set according to the actual climbing trajectory of the slipform system during the construction of the target cylinder; The initial axis is optimized by iterative reweighted least squares method to obtain the current center axis of the target cylinder; wherein the weights of the iterative reweighted least squares method are dynamically adjusted according to the reciprocal of the average distance between the point cloud in each cross-sectional plane and the initial axis.

7. The method for visual monitoring of the construction posture of a cylindrical structure based on three-dimensional modeling as described in claim 1, characterized in that, The construction of a digital twin model of the target cylinder based on the three-dimensional point cloud data includes: The three-dimensional point cloud data is segmented to obtain the cylindrical structural components of the target cylinder represented by the point cloud. Obtain the semantic information of the preset BIM model of the target cylinder; establish the association mapping between the cylinder structural components and the semantic information respectively; Based on the association mapping and the preset BIM model, a semantic digital model containing the topological connection relationships of components is constructed; The semantic digital model is updated by collecting construction progress data in real time to obtain the digital twin model.

8. A visualization monitoring system for the construction posture of a cylindrical structure based on 3D modeling, characterized in that, It includes a model building module, a planned data acquisition module, a centerline analysis module, an offset vector calculation module, and a visualization and monitoring module; among which, The model building module is used to collect three-dimensional point cloud data of the target cylinder and build a digital twin model of the target cylinder based on the three-dimensional point cloud data. The planning data acquisition module is used to filter out key surface points of the target cylinder based on the digital twin model; and acquire construction plan data of the target cylinder; wherein, the construction plan data includes preset cylinder model data, and the preset cylinder model data includes design axis and joint parts; The central axis analysis module is used to analyze and obtain the current central axis of the target cylinder based on the surface key points; and to analyze and obtain the design coordinates of each surface key point in the preset cylinder model based on the preset cylinder model data based on the selected surface key points. The offset vector calculation module is used to calculate the current first offset vector based on the central axis and the design axis; and to calculate the current second offset vector of each surface key point according to the current coordinates and design coordinates of the surface key points. The visualization monitoring module is used to determine the first offset vector as an abnormal vector when preset conditions are met; based on the abnormal vector, the second offset vector, and the position of the seam, select the seam key points affected by misalignment from the surface key points; and annotate each of the seam key points in the digital twin model, visualize the annotated digital twin model, and monitor the annotated digital twin model.

9. The visualization monitoring system for the construction posture of a cylindrical structure based on three-dimensional modeling as described in claim 8, characterized in that, The visualization monitoring module, based on the second offset vector and the position of the seam, filters out the seam key points affected by misalignment from the surface key points, including: The visualization monitoring module filters out an initial set of seam points from key surface points that are within a preset range of distance from the seam, based on the location of the seam. Calculate the radial and axial components of the second offset vector for each point in the initial joint point set; wherein the radial component is perpendicular to the design axis and the axial component is parallel to the design axis; Based on the distribution of the radial components, the radial misalignment region of the joint is determined by cluster analysis; wherein, the radial misalignment region is a continuous set of points where the radial component value exceeds a first preset threshold. Based on the distribution of the axial components, the axial misalignment region of the joint is determined by cluster analysis; wherein, the axial misalignment region is a continuous set of points where the axial component value exceeds a second preset threshold. The initial joint point located in the radial misalignment region or the axial misalignment region is determined as the joint critical point.

10. The visualization monitoring system for the construction posture of a cylindrical structure based on three-dimensional modeling as described in claim 8, characterized in that, The central axis analysis module analyzes and obtains the current central axis of the target cylinder based on the surface key points, including: The central axis analysis module establishes an initial reference coordinate system based on the direction of the design axis of the target cylinder, projects each surface key point onto the initial reference coordinate system, and the projection results of each surface key point are perpendicular to multiple equally spaced cross-sectional planes of the design axis, thereby obtaining the projection points corresponding to each surface key point; Within each of the cross-sectional planes, the projection points are fitted with an ellipse using a random sampling consensus algorithm to obtain the fitting result; wherein, the number of samplings in the random sampling consensus algorithm is adaptively adjusted based on the distribution entropy value of the point cloud within the cross-sectional plane; Based on the fitting results, the spatial continuity between the elliptical centers of adjacent cross-sectional planes is calculated; and a smoothness constraint for the center point's motion trajectory is constructed based on the spatial continuity. Based on the smoothness constraints, the current central axis of the target cylinder is obtained according to the centers of each ellipse.

Images