A steel plant pipeline digital model fitting method based on a mobile three-dimensional laser scanner

Abstract

The application discloses a steel plant pipeline digital model fitting method based on a mobile three-dimensional laser scanner. The method comprises the following steps: step one, a laser scanner is used to scan a target pipeline to collect original pipeline point cloud data; step two, space analysis is performed on the point cloud data of the pretreated straight pipe to extract a main direction vector of the straight pipe; step three, the straight pipe point cloud is divided by slicing based on the extracted main direction vector, and a corresponding circle center position and radius value are fitted; step four, for the elbow pipe part, the slice circle center where the straight pipe and the elbow pipe meet is taken as the boundary of the elbow pipe; the center point and the radius of the elbow pipe section are determined; step five, the pipeline is geometrically modeled in combination with the center coordinates and the radius information of each pipe section, and a three-dimensional geometric model conforming to the actual pipeline is generated. The application restores the geometric characteristics of the actual pipeline by slicing fitting and center line reconstruction.

Description

Technical Field

[0001] This invention relates to a method for fitting digital models of steel plant pipelines based on a mobile 3D laser scanner. Background Technology The utilities and process piping networks in steel plant areas are vast, complex, and densely intersecting. They have also undergone frequent renovations, maintenance, and temporary rerouting, leading to a common discrepancy between the actual site and the blueprints. Existing as-built drawings, process flow diagrams, and BIM models often fail to reflect newly added, scrapped, or relocated pipe sections in a timely manner. This discrepancy directly impacts subsequent operation management, safety and environmental assessments, emergency repair location, and the accuracy and usability of digital twin platforms.

[0002] The emergence of 3D laser scanning technology has provided a new approach to pipeline modeling. By scanning the target scene, a large amount of spatial point cloud data can be acquired in a short time, providing data support for subsequent geometric analysis and modeling. However, raw point cloud data often contains noisy points, outliers, and redundant information. If modeling is carried out directly without processing, it can easily lead to insufficient model accuracy or morphological distortion. Therefore, how to balance efficiency and accuracy in the preprocessing, orientation extraction, and geometric fitting of point cloud data has become an important research direction in the field of pipeline modeling. Summary of the Invention

[0003] To address the problems existing in the prior art, this invention provides a method for fitting digital models of steel plant pipelines based on a mobile 3D laser scanner, thereby enabling the reconstruction of pipeline models using 3D point cloud data.

[0004] To achieve the above objectives, the present invention provides a method for fitting a digital model of a steel plant pipeline based on a mobile 3D laser scanner, comprising the following steps: Step 1: Use a laser scanner to scan the target pipe, collect the original pipe point cloud data, and preprocess the point cloud data; Step 2: Perform spatial analysis on the preprocessed point cloud data of the straight pipe to extract the principal direction vector of the straight pipe; Step 3: Based on the extracted main direction vector, the straight tube point cloud is sliced ​​and the point set is determined, and the corresponding circle center position and radius value are obtained by fitting. Step 4: For the bend section, take the center of the slice where the straight pipe meets the bend as the boundary of the bend; and determine the center point and radius of the bend section. Step 5: Combine the center coordinates and radius information of each pipe segment to perform overall geometric modeling of the pipeline, and then generate a three-dimensional geometric model that conforms to the actual pipeline.

[0005] Furthermore, the method for generating the three-dimensional geometric model includes the following steps: Center of the slicec k Connect the sequences sequentially and perform curve fitting to generate the three-dimensional centerline of the pipeline. C ( s ); The radius obtained by fitting the slice { r k Interpolation is performed to obtain a continuous radius function distributed along the centerline. r ( s (), used to describe the continuous variation of the pipe radius with respect to its length; Based on the three-dimensional centerline C ( s ) and radius distribution function r ( s Generate a complete 3D table of pipe start point, end point, and pipe diameter information, and use Dynamo to achieve one-click pipe modeling in Revit.

[0006] Furthermore, the pipe slice division and point set determination in step three include the following steps: 3.1 Point Cloud p i (0 < i ≤ N (N is the number of point clouds in the current straight pipe) in the spatial direction of the straight pipe Projecting onto the pipe, we obtain the projected coordinates along the pipe axis: ; 3.2 Determine the range based on the projected coordinates [ t min ,t max The pipeline axis is divided into... K part; 3.3 Point Cloud Aggregate P =[ p 1, p 2,…, p i The projected coordinates of the point cloud along the pipeline axis are calculated as follows: T =[ t 1, t 2,…, t i ],according to K Segmentation set T The dividing boundary is edges =[ e 0, e 1,…, e K ]; 3.4 The selection rule is as follows: For each point in the point cloud set, a total of [number] points are selected. KThis process involves determining the slice to which each point belongs, ensuring that the number of points within each slice is not less than a threshold. k min The slice is then retained, and the slice point is determined as follows:

[0007]

[0008] The final set of slices is: Extract each segment of the point cloud as a slice point set. S k (1≤ k ≤ K ).

[0009] Furthermore, determining the center point of the bend in step four includes the following steps: 4.1 Cut the center of the connected straight pipe sections into circles. c 0 as the initial point and the first sliding window W The center of the sphere is set to the radius of the adjacent straight pipe segment slices. R ; 4.2 Based on the aforementioned initial point c 0 and neighborhood radius R The following steps 4.3 to 4.4 are performed iteratively. 4.3 For the first n The next iteration, with the current window c n For the center of the ball, R With radius, point cloud from the bend section P’ Extract the points located within the spherical neighborhood to form the current window point set. W n ; 4.4 Calculate the window point set W n center of mass p n The covariance matrix, after eigenvalue decomposition, satisfies the following eigenvalues: Their corresponding unit eigenvectors are respectively , , Get the centerline point of the current window q n : ,in: Add this point to the center point sequence C middle; 4.5 Set the center point of the current window q n Along the largest eigenvalue Corresponding feature vector Move one step in the direction Step Get the center point of the next window. c k+1 : ,make n = n +1, repeat steps 4.3 to 4.5 until the iteration termination condition is met.

[0010] Furthermore, the preprocessing includes removing outliers and noise points, and downsampling the point cloud to improve the efficiency and accuracy of subsequent calculations.

[0011] This invention employs mobile 3D laser scanning technology to rapidly acquire large-scale pipeline point cloud data, automating and maximizing the efficiency of pipeline modeling. Through point cloud preprocessing and pipeline orientation extraction based on principal component analysis, the impact of noise interference and redundant information is effectively reduced. By slicing and centerline reconstruction, the geometric features of the actual pipeline are realistically reproduced. The generated pipeline model provides a reliable data foundation for digital twin systems, supporting pipeline operation management, health monitoring, maintenance, and repair. Attached Figure Description

[0012] Figure 1 This is a technical flowchart of the present invention.

[0013] Figure 2 This is a schematic diagram of a straight tube slice and its fitting.

[0014] Figure 3 It is an instance point cloud data map.

[0015] Figure 4 It is a preprocessed point cloud data map.

[0016] Figure 5 This is a slice diagram of the straight pipe section in the example.

[0017] Figure 6 It is the spherical neighborhood graph of the bend in the example.

[0018] Figure 7 Dynamo pipeline modeling procedure diagram.

[0019] Figure 8 A diagram illustrating an example of pipeline modeling in this invention. Detailed Implementation

[0020] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0021] In the description of this invention, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0022] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0023] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0024] like Figures 1 to 8 As shown, the method for fitting digital models of steel plant pipelines based on a mobile 3D laser scanner of the present invention is applicable to the construction of 3D models of factories and other structures containing pipelines, and includes the following steps: Step 1: Use a mobile 3D laser scanner to scan the target pipeline, collect the original pipeline point cloud data, and preprocess the point cloud data, including removing outliers and noise points, and downsampling the point cloud to improve the efficiency and accuracy of subsequent calculations.

[0025] Step 2: Perform spatial analysis on the preprocessed straight pipe point cloud data to extract the main direction vector of the straight pipe. Step 3: Based on the principal direction vector, the straight tube point cloud is divided into slices and the point set is determined. RANSAC fitting is performed on each slice of point cloud to obtain the corresponding center position and radius value. This includes the following sub-steps: 3.1 Point Cloud p i (0 < i ≤ N (N is the number of point clouds in the current straight pipe) in the spatial direction of the straight pipe Projecting onto the pipe, we obtain the projected coordinates along the pipe axis: ; 3.2 Determine the range based on the projected coordinates [ t min ,t max The pipeline axis is divided into... K part; 3.3 Point Cloud Aggregate P =[ p 1, p 2,…, p i The projected coordinates of the point cloud along the pipeline axis are calculated as follows: T =[ t 1, t 2,…, t i ],according to K Segmentation set T The dividing boundary is edges =[ e 0, e 1,…, e K ]; 3.4 The selection rule is as follows: For each point in the point cloud set, a total of [number] points are selected. K This process involves determining the slice to which each point belongs, ensuring that the number of points within each slice is not less than a threshold. k min The slice is then retained, and the slice point is determined as follows:

[0026]

[0027] The final set of slices is: Extract each segment of the point cloud as a slice point set. S k (1≤ k ≤ K ).like Figure 2 The image shows a slice of a straight tube and a schematic diagram of the fitting process.

[0028] Step 4: For the bend, the center of the slice where the straight pipe meets the bend is taken as the boundary of the bend. Using a sliding window and local principal component analysis, the center point of the bend segment is determined, and the points within the sliding window are fitted with RANSAC to obtain the radius value. This includes the following sub-steps: 4.1 Cut the center of the connected straight pipe sections into circles. c 0 as the initial point and the first sliding window W The center of the sphere is set to the radius of the adjacent straight pipe segment slices. R ; 4.2 Based on the aforementioned initial point c 0 and the radius of the domainR The following steps 3.3 to 3.4 are executed iteratively. 4.3 For the first n The next iteration, with the current window c n For the center of the ball, R With radius, point cloud from the bend section P’ Extract the points located within the spherical neighborhood to form the current window point set. W n ; 4.4 Calculate the window point set W n center of mass p n The covariance matrix, after eigenvalue decomposition, satisfies the following eigenvalues: Their corresponding unit eigenvectors are respectively , , Get the centerline point of the current window q n : ,in: Add this point to the center point sequence C middle; 4.5 Set the center point of the current window q n Along the largest eigenvalue Corresponding feature vector Move one step in the direction Step Get the center point of the next window. c k+1 : ,make n = n +1, repeat steps 4.3 to 4.5 until the iteration termination condition is met.

[0029] Step 5: Combining the center coordinates and radius information of each pipe segment, perform overall geometric modeling of the pipeline to generate a 3D geometric model that conforms to the actual pipeline. This includes the following sub-steps: 5.1 Center the slice c k Connect the sequences sequentially and perform curve fitting to generate the three-dimensional centerline of the pipeline. C ( s ); 5.2 The radius obtained by fitting the slice { r k Interpolation is performed to obtain a continuous radius function distributed along the centerline. r ( s (), used to describe the continuous variation of the pipe radius with respect to its length; 5.3 Based on the aforementioned three-dimensional centerline C ( s ) and radius distribution function r ( s Generate a complete 3D table of pipe start point, end point, and pipe diameter information, and use Dynamo to achieve one-click pipe modeling in Revit.

[0030] Example 1 This embodiment of a method for fitting steel plant pipes based on a mobile 3D laser scanner includes the following steps: Step 1: Use a mobile 3D laser scanner to scan the target pipe and acquire the original pipe point cloud data, such as... Figure 3 As shown, the original points contain a certain degree of noise. This example uses Python and the Open3D library to process point cloud data acquired by a mobile 3D laser scanner. The Open3D library reads the scanner's PLY format point cloud file, and voxel downsampling is used to reduce the number of points and improve subsequent computational efficiency. A statistical outlier removal algorithm is then used to remove isolated and noisy points. The processed pipeline point cloud is shown below. Figure 4 As shown.

[0031] Step 2: Perform spatial analysis on the preprocessed straight pipe point cloud data to extract the principal direction vector of the straight pipe; Step 3: Based on the principal direction vector, slice the point cloud, project the point cloud onto the spatial direction of the pipe to obtain the projected coordinates along the pipe axis, slice the coordinate range and obtain the points within the slices, and use RANSAC fitting to obtain the corresponding center position and radius value. A schematic diagram of the slicing division of the straight pipe section is shown below. Figure 5 As shown; Step 4: For the bend, the center of the slice where the straight pipe meets the bend is taken as the boundary of the bend. Using a sliding window and local principal component analysis, the center point of the bend segment is determined. The points within the sliding window are then fitted with RANSAC to obtain the radius value. A schematic diagram of the point set within a spherical sliding window of the bend is shown below. Figure 6 As shown; Step 5: Combining the center coordinates and radius information of each pipe segment, perform overall geometric modeling of the pipeline to generate a 3D geometric model that matches the actual pipeline. Export the pipeline start-point, end-point, and diameter information to an Excel file. Use the "File path" node in Dynamo to create a file path object, use the "File From Path" node to obtain the Excel file entity from the path, and use the "Data.ImportExcel" node to read data from the Excel file. Use List.AllIindicesOf, List.GetItemAtIndex, Point.ByCoordinates, and List.Create to construct the coordinate information of the pipeline start-point and end-point. Select the pipeline family established by the adaptive metric conventional model in Family Types. Use these two parts as inputs to AdaptiveComponent.ByPoints, and use the Element.SetParameterByName node to update in real time from Dynamo based on the pipeline and diameter information. After importing the program into the local Dynamo player and inputting the pipeline parameter table, one-click modeling is complete.

[0032] The present invention has been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the embodiments described above. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention. Many other changes and modifications made without departing from the concept and scope of the present invention should be considered within the scope of protection of the present invention.

[0033] In the description of this specification, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0034] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for fitting digital models of steel plant pipelines based on a mobile 3D laser scanner, characterized in that, Includes the following steps: Step 1: Use a laser scanner to scan the target pipe, collect the original pipe point cloud data, and preprocess the point cloud data; Step 2: Perform spatial analysis on the preprocessed point cloud data of the straight pipe to extract the principal direction vector of the straight pipe; Step 3: Based on the extracted main direction vector, the straight tube point cloud is sliced ​​and the point set is determined, and the corresponding circle center position and radius value are obtained by fitting. Step 4: For the bend section, take the center of the slice where the straight pipe meets the bend as the boundary of the bend; and determine the center point and radius of the bend section. Step 5: Combine the center coordinates and radius information of each pipe segment to perform overall geometric modeling of the pipeline, and then generate a three-dimensional geometric model that conforms to the actual pipeline.

2. The method for fitting a digital model of a steel plant pipeline based on a mobile 3D laser scanner as described in claim 1, characterized in that, The method for generating the three-dimensional geometric model includes the following steps: Center of the slice c k Connect the sequences sequentially and perform curve fitting to generate the three-dimensional centerline of the pipeline. C ( s ); The radius obtained by fitting the slice { r k Interpolation is performed to obtain a continuous radius function distributed along the centerline. r ( s (), used to describe the continuous variation of the pipe radius with respect to its length; Based on the three-dimensional centerline C ( s ) and radius distribution function r ( s Generate a complete 3D table of pipe start point, end point, and pipe diameter information, and use Dynamo to achieve one-click pipe modeling in Revit.

3. The method for fitting a digital model of a steel plant pipeline based on a mobile 3D laser scanner as described in claim 1, characterized in that, The pipe slice division and point set determination in step three include the following steps: 3.1 Point Cloud p i (0 < i ≤ N (N is the number of point clouds in the current straight pipe) in the spatial direction of the straight pipe Projecting onto the pipe, we obtain the projected coordinates along the pipe axis: ; 3.2 Determine the range based on the projected coordinates [ t min ,t max The pipeline axis is divided into... K part; 3.3 Point Cloud Aggregate P =[ p 1, p 2,…, p i The projected coordinates of the point cloud along the pipeline axis are calculated as follows: T =[ t 1, t 2,…, t i ],according to K Segmentation set T The dividing boundary is edges =[ e 0, e 1,…, e K ]; 3.4 The selection rule is as follows: For each point in the point cloud set, a total of [number] points are selected. K This process involves determining the slice to which each point belongs, ensuring that the number of points within each slice is not less than a threshold. k min The slice is then retained, and the slice point is determined as follows: The final set of slices is: Extract each segment of the point cloud as a slice point set. S k (1≤ k ≤ K ).

4. The method for fitting a digital model of a steel plant pipeline based on a mobile 3D laser scanner as described in claim 1, characterized in that, Determining the center point of the bend in step four includes the following steps: 4.1 Cut the center of the connected straight pipe sections into circles. c 0 as the initial point and the first sliding window W The center of the sphere is set to the radius of the adjacent straight pipe segment slices. R ; 4.2 Based on the aforementioned initial point c 0 and neighborhood radius R The following steps 4.3 to 4.4 are performed iteratively. 4.3 For the first n The next iteration, with the current window c n For the center of the ball, R With radius, point cloud from the bend section P’ Extract the points located within the spherical neighborhood to form the current window point set. W n ; 4.4 Calculate the window point set W n center of mass p n The covariance matrix, after eigenvalue decomposition, satisfies the following eigenvalues: Their corresponding unit eigenvectors are respectively , , Get the centerline point of the current window q n : ,in: Add this point to the center point sequence C middle; 4.5 Set the center point of the current window q n Along the largest eigenvalue Corresponding feature vector Move one step in the direction Step Get the center point of the next window. c k+1 : ,make n = n +1, repeat steps 4.3 to 4.5 until the iteration termination condition is met.

5. The method for fitting a digital model of a steel plant pipeline based on a mobile 3D laser scanner as described in claim 1, characterized in that, The preprocessing includes removing outliers and noise points, and downsampling the point cloud to improve the efficiency and accuracy of subsequent calculations.

Images