Novel ship building forming component three-dimensional design software and curved surface forming detection interface method

Abstract

The invention discloses a novel three-dimensional design software and curved surface forming detection interface method for a ship building forming component, and belongs to the technical field of ship curved surface three-dimensional design. The method comprises the following steps: generating a curved plate boundary and internal points according to curved plate information extracted from ship segmentation model data to generate an NURBS curve; generating an NURBS curve according to welding baseline discrete points of a curved profile extracted from a ship segmentation model; intercepting the NURBS curve generated by the discrete points of the welding base line according to the plane plate information extracted by the ship segmentation model; performing point set encryption according to the intercepted NURBS curve of the welding base line to generate an intersected NURBS curve, and generating an NURBS grid of the curved plate; and performing NURBS curved surface reconstruction according to the intercepted NURBS curve of the welding base line. According to the method, ship segmented curved surface plate reconstruction is carried out based on NURBS curve curved surface reconstruction, three-dimensional design software and a curved surface forming detection structure method of ship construction forming components can be formed, and curved surface plate reconstruction exported from ship production design software is achieved.

Description

Technical Field

[0001] The invention relates to the field of three-dimensional design of ship curved surfaces, and in particular to a three-dimensional design software for shipbuilding molding components and a curved surface molding detection interface method. Background Art

[0002] During the shipbuilding process, the forming components of the hull structure are usually modeled and designed using 3D design software. These design data need to be passed to the subsequent forming inspection link to ensure that the processing and installation of the components meet the design requirements. However, there are many problems with the data interaction between the existing design software and the forming inspection system, which are mainly manifested in the following aspects:

[0003] a. Incompatible data formats: Different software systems often use different data formats, resulting in data loss or format errors during data transmission.

[0004] b. Inaccurate surface forming detection: When performing surface forming error detection, the existing detection system is difficult to accurately match the design data, which affects the accuracy of the detection results.

[0005] c. Lack of automated interface: In the existing technology, the interface between design data and test data is mostly manual operation, which is inefficient and prone to errors.

[0006] Therefore, there is an urgent need for a new method that can realize data transmission and interface docking between the 3D design software of shipbuilding forming components and the surface forming detection system, improve work efficiency and ensure the precise matching of design and forming.

[0007] Based on the above technical problems, from the perspective of solving different technical problems, the following multiple documents can also be detected, but they also have various problems, as follows:

[0008] Patent CN116091734A, a method for characterizing geometric errors based on NURBS surface reconstruction; the technical point of this patent is to determine the degree of the NURBS surface and the weight factor of each shape value point; to parameterize the shape value points in the row direction u and the column direction v respectively using the averaging technique AVG; to perform NURBS curve interpolation to obtain the coordinates of all control vertices; to use NURBS surface interpolation and reconstruction technology to establish a mathematical model of the true geometric shape error of the surface of the part to be processed, and to generate a geometric error surface model through direct calculation.

[0009] Patent CN110796735A, NURBS surface finite element plate and shell meshing method and computer implementation system; the technical point of this patent is to obtain the original NURBS surface, select multiple seed points on the boundary of the original NURBS surface; map the natural coordinates of the seed points in the surface to the plane, and perform triangular meshing on the plane based on the seed points.

[0010] Paper [1] Li Chuanjun, Wang Liping. Research on B-spline curve fitting algorithm for airfoil blade-like surfaces. Computer Integrated Manufacturing Systems, 2024, 30(01): 144-157. The technical point of this paper is to reconstruct the discrete tool path based on fluid mechanics characteristics using B-spline curves, and fit the equivalent lift tool trajectory under the conditions of minimum curvature segmentation, endpoint interpolation, endpoint tangent vector continuity and maximum allowable error.

[0011] Paper [2] Liu Xinkai. Research on trajectory planning of large ship facade spraying robot based on parametric surface reconstruction. Southeast University, 2022. The technical point of this paper is to pre-segment the ship facade according to the process requirements of ship segment painting and reconstruct the NURBS surface of the pre-segmented ship facade.

[0012] Paper [3] Zhang Xu, Hou Maosheng, Liu Zhichao, et al. Surface reconstruction algorithm for plate and shell structures based on fiber Bragg grating sensors. Laser and Optoelectronics Progress, 2020, 57(09): 74-81. The technical point of this paper is to use fiber Bragg grating (FBG) sensors to study the surface reconstruction algorithm for plate and shell structures in order to solve the assembly deformation problem caused by factors such as prestress and dimensional position deviation. The relationship between wavelength offset and curvature is established, and the curvature data required for the surface reconstruction algorithm is obtained. The segmented fitting algorithm is used to calculate the coordinate increments of all measurement points, thereby realizing curve reconstruction.

[0013] In summary, the existing technology usually focuses on the generation of surfaces, and there is little research on the reconstruction of surfaces derived from production design software for the segmented curved plate structure of a ship. Summary of the invention

[0014] In view of the deficiencies in the prior art, the present invention proposes a novel three-dimensional design software for shipbuilding forming components and a curved surface forming detection interface method. By analyzing the curved plate boundary point data and internal point data, the profile welding baseline discrete point data on the curved plate, and the plane plate boundary point discrete data in the AM-XML file, a novel three-dimensional design software for shipbuilding forming components and a curved surface forming detection interface method is proposed. The method innovatively proposes a NURBS curve reconstruction method based on the profile welding baseline discrete data, and designs a bounding box according to the curved plate boundary discrete data, and intercepts the NURBS curve reconstructed by the profile welding baseline discrete data to generate a real profile welding baseline NURBS curve attached to the curved plate.

[0015] The three-dimensional design software for shipbuilding forming components and the curved surface forming detection interface method of the present invention include the following steps:

[0016] Step 1: Methods for structured data analysis and database creation

[0017] Extract data from the AM-XML file and create a database. Use the segment name as the database name and the curved plate name as the sub-database name to implement structured data parsing and database creation.

[0018] Step 2: Data parsing and topological analysis method for multi-subsurface plates

[0019] The curved plate data is parsed, the boundary points and internal points of the curved plate are extracted, and the bounding box of the curved plate is constructed. The bounding box is used to judge the boundary point data of the flat plate and analyze the topological relationship of the curved plate.

[0020] Step 3: Profile welding baseline NURBS curve reconstruction process

[0021] The NURBS curve is reconstructed for the discrete point data of the profile welding baseline and saved in the database file.

[0022] Step 4: Optimize the profile welding baseline NURBS curve, encrypt the NURBS curve discrete point set, and generate curve family III

[0023] The profile welding baseline NURBS curve is intercepted by a bounding box and encrypted to generate an encrypted point set. The encrypted point set is analyzed to generate a new curve family III perpendicular to the profile welding baseline NURBS curve, so that the curved plate presents a grid-like distribution.

[0024] Step 5: Generate NURBS surface for curved plate

[0025] The surface is reconstructed using the intercepted profile welding baseline NURBS curve.

[0026] Furthermore, the step 2 of solving the multi-sub-surface plate data parsing and topological analysis method comprises:

[0027] The boundary point information in the curved plate data is identified and judged to determine the boundary point information used for NURBS surface generation; and the boundary point information in the plane plate data is topologically analyzed to determine the connected plane plates.

[0028] Furthermore, the step S2 specifically includes the following sub-steps:

[0029] S2.1 Data extraction and storage

[0030] Further analyzing multiple sub-curved plate materials for a certain curved plate material corresponding to the profile welding baseline in step S1; and extracting node data, each sub-curved plate material contains k curves, and each curve corresponds to g spatial points;

[0031] Save the point data of each sub-surface plate, and the coordinates of the points are expressed as:

[0032] ;

[0033] S2.2 generates a bounding box and parses the relevant information for each point of the sub-surface plate:

[0034]

[0035] Create a bounding box based on the above range and name it: ;

[0036] S2.3 Generate boundary lines, generate boundary lines of curved plates based on point data :

[0037]

[0038]

[0039]

[0040]

[0041] S2.4 Topological analysis of planar plates

[0042] For the nodes in the AM-XML file, extract the boundary points of the plane plate; for a plane plate boundary point (x, y, z), determine whether it is within the bounding box:

[0043]

[0044] If the conditions are met, it is determined that there is a topological association between the plane plate and the sub-curved plate.

[0045] S2.5 Calculation of base plane of planar plate

[0046] For the topologically associated plane plate boundary point set , calculate the following parameters:

[0047]

[0048] Determine the minimum value , determine the base plane of the planar plate:

[0049] like Minimum, the base plane is YOZ;

[0050] like Minimum, the base plane is XOZ;

[0051] like Minimum, base plane is XOY.

[0052] Furthermore, the step S3 specifically includes the following sub-steps:

[0053] Based on profile welding baseline discrete point data , using the cumulative chord length method to calculate non-uniform node vectors;

[0054] S3.1 For a given point set , calculate the Euclidean distance between each pair of adjacent points: ;

[0055] S3.2 Calculate the cumulative chord length, cumulative chord length Defined as: ,final is the total chord length;

[0056] S3.3 Normalize the parameter value, normalize the cumulative chord length, and map it to the interval [0, 1]:

[0057]

[0058] S3.4 constructs the node vector. For a NURBS curve with a curve order of p, the node vector Constructed as: Repeat at the beginning and end times, that is The intermediate node values ​​are all parameter values , but remove the first and last ends;

[0059] S3.5 Obtained by the cumulative chord length method After constructing the parameter values, calculate the basis function matrix ; Compute B-spline basis functions using recursive definition , construct the matrix :

[0060]

[0061] S3.6 Solve the linear equations. Solve the linear equations according to the interpolation conditions. ,in is a column vector of control points, if For square ;if It is not a square matrix, and the least squares solution is required. ;

[0062] S3.7 generates the profile welding baseline according to the NURBS curve formula:

[0063]

[0064] According to the above NURBS curve definition and formula, the NURBS curve formula of each profile welding baseline can be calculated.

[0065] Furthermore, using the bounding box Intercepting the multiple profile welding baselines in step S3 to obtain multiple new profile welding baselines;

[0066] Determine the "X", "Y", and "Z" values ​​of discrete points on the profile welding baseline and solve the corresponding parameters , is the difference between the maximum and minimum values ​​of "X", is the difference between the maximum and minimum values ​​of "Y", is the difference between the maximum and minimum values ​​of the "Z" value; The minimum value in The minimum, then the profile welding baseline is parallel to the YOZ plane, that is, the base plane of the profile welding baseline is the YOZ plane; assuming The minimum, then the profile welding baseline is parallel to the XOZ plane, that is, the base plane of the profile welding baseline is the XOZ plane; assuming The minimum, then the profile welding baseline is parallel to the XOY plane, that is, the base plane of the profile welding baseline is the XOY plane;

[0067] Combined with the base plane of the profile welding baseline, the NURBS curve corresponding to the boundary line is filled; the discrete point data of the profile welding baseline and the boundary line are arranged from small to large in Z value, and then the curve family I of the curved plate is generated;

[0068] For multiple curves in the curve family I, the points are refined. Assuming the step size is λ, the "X" value of the point where the longitudinal section line is refined is: X=Xmin+λs and Xmin≤X≤Xmax, s=0,1,…; each NURBS curve in the curve family I is known. When the X value is obtained, the points on each NURBS curve can be refined; that is, the formula of the known NURBS curve is converted, and the X value of a point on the curve is known, the coordinate problem of the point can be solved;

[0069] The encrypted points and bounding box Make a judgment, assuming that the encrypted point is within the bounding box, then keep the encrypted point. The rest are deleted, and then the encrypted points of the profile that exceed the boundary of the curved plate are deleted; then the profile welding baseline NURBS curve is regenerated; and the NURBS curve corresponding to the curve corresponding to the boundary line is merged to form curve family II;

[0070] By connecting corresponding points on multiple NURBS curves with the same X value, discrete curve points perpendicular to multiple curves on curve family II are regenerated, and multiple curves of curve family III are reconstructed using the method in step S3.

[0071] Furthermore, the sub-steps of step S5 are as follows:

[0072] S5.1 NURBS curve extraction and information acquisition: For each NURBS curve in curve family II, extract its control points, weights and node vector information in turn to construct a complete NURBS parametric description;

[0073] S5.2 Multi-curve fitting based on B-spline: Construct a B-spline mesh composed of curves and map the curve data of curve family II into the surface fitting framework;

[0074] Using the weighted average of the control points and the geometric constraints between the curves, a B-spline surface that conforms to the curve family II data is preliminarily generated; the constraints include the continuity between the curves and the smoothness of the spatial distribution of the control points;

[0075] S5.3 Surface fitting optimization: Use fitting algorithms to optimize the surface and reduce the error between the fitted surface and the original curve family data; adjust the control point position, weight distribution and node vector configuration to optimize the surface shape and make it closer to the geometric characteristics of the original data;

[0076] Recalculate the error index for each curve on the fitting surface to ensure that the deviations of all control points and specified sampling points are within the threshold range;

[0077] S5.4 Dynamic adjustment and error evaluation: After the surface is generated, the control points, weights and node vector distribution are further adjusted through iterative optimization; combined with the error evaluation method, the fitting process is dynamically optimized;

[0078] S5.5 Surface data storage and model reconstruction: The final optimized B-spline surface parameter description: control points, weights, node vectors, surface equations are stored in the database to form standardized data records.

[0079] The beneficial effects of the present invention are:

[0080] The method generates NURBS curves based on the curved plate information extracted from the ship segment model data, generates NURBS curves based on the boundary and internal points of the curved plate, generates NURBS curves based on the welding baseline discrete points of the curved profile extracted from the ship segment model; intercepts the NURBS curves generated by the welding baseline discrete points based on the plane plate information extracted from the ship segment model; generates intersecting NURBS curves based on the intercepted welding baseline NURBS curve, and generates a NURBS grid of the curved plate; reconstructs the NURBS surface based on the intercepted welding baseline NURBS curve. The present invention reconstructs the curved plate of the ship segment based on the reconstruction of NURBS curves and surfaces, and can form a three-dimensional design software for ship construction forming components and a curved surface forming detection structure method, and realizes the reconstruction of the curved plate derived from the ship production design software. The method effectively solves the problems of large fitting errors and low detection accuracy caused by insufficient discrete points of the curved plate solved by the line diagram and insufficient discrete data points of the curved plate extracted from the AM model in the current technology.

[0081] This method innovatively proposes a NURBS curve reconstruction method based on the discrete data of the profile welding baseline, designs a bounding box based on the discrete data of the curved plate boundary, and intercepts the NURBS curve reconstructed from the discrete data of the profile welding baseline to generate a real profile welding baseline NURBS curve attached to the curved plate. It innovatively proposes to use the profile welding baseline NURBS curve to generate encrypted point sets, and generates NURBS curves twice by using the encrypted point set data on multiple NURBS curves, so that the two sets of NURBS curves generate mesh information. Finally, the profile welding baseline NURBS curve and the curved plate boundary NURBS curve are innovatively used to reconstruct the surface. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0083] Figure 1 It is a flow chart of the three-dimensional design software for shipbuilding forming components and the curved surface forming detection structure method in an embodiment of the present invention.

[0084] Figure 2 Welding baseline for profiles in AM-XML files.

[0085] Figure 3 The surface mesh fitted by curve family II and curve family III.

[0086] Figure 4 The surface fitted by curve family II. DETAILED DESCRIPTION

[0087] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technology in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than comprehensive embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0088] Figure 1 The flowchart of the novel three-dimensional design software and curved surface forming detection structure method for shipbuilding forming components in the present invention is as follows: Figure 1 As shown, the method comprises the following steps:

[0089] Step 1: Based on the XML file of the ship segment AM model, a method of structured data parsing and database creation is proposed to efficiently manage the information of the ship segment surface structure. Specifically, it includes:

[0090] Based on the XML file of the AM model of ship sections (hereinafter referred to as AM-XML file), a method of structured data parsing and database creation is proposed to efficiently manage the hull section construction information. The specific steps are as follows:

[0091] Step 101. Determine segment name and create database

[0092] The attribute "ObjId" of the "Block" subnode under the "Ship" node in the AM-XML file is used as the unique identifier of the segment to determine the segment name.

[0093] A corresponding database is established according to the segment name to store all relevant data of the segment.

[0094] Step 102. Determine the name of the curved plate and create a sub-database

[0095] The attribute “GroupId” of the “CurvedPanel—PlateGroup” subnode under the “Block” subnode is used as the unique identifier of the curved plate material to determine the name of the curved plate material.

[0096] An independent sub-database is established for each curved plate to facilitate the classification and management of the curved plate related data within the segment.

[0097] Step 103. Extraction of profile data

[0098] For the database created above, parse the profile information under each "Ship—Block—CurvedPanel" node.

[0099] Data is extracted one by one for multiple profile nodes in "Ship—Block—CurvedPanel—StiffenerGroup", including key data of profile welding baseline.

[0100] Step 104. Analysis of profile welding baseline data

[0101] Extract the data in "Ship—Block—CurvedPanel—StiffenerGroup—Stiffener--Trace" as the core of the profile welding baseline data.

[0102] Focus on extracting the discrete point data in "StartPoint2d" and "Segment2d—Node2d" under the "Trace" subnode to ensure the integrity of the geometric information of the welding baseline.

[0103] In the curved panel (“Ship—Block—CurvedPanel”), each profile welding baseline forms a continuous curve in three-dimensional space through its discrete point data. Assume that there is The welding baseline of each profile is arranged in sequence as a curve in three-dimensional space.

[0104] Step 2: Propose a method for data parsing and topological analysis of multi-sub-surface plates. Specifically, it includes:

[0105] Step 201. Data extraction and storage

[0106] For a curved plate ("Ship—Block—CurvedPanel") corresponding to the profile welding baseline in step 1, multiple sub-curved plates ("Ship—Block—CurvedPanel—PlateGroup") are further parsed.

[0107] Extract the data of the "Ship—Block—CurvedPanel—PlateGroup—Plate—FaceSurface" node. Each sub-surface plate contains 9 curves, and each curve corresponds to 9 spatial points.

[0108] Save the 81 point data of each sub-surface plate, and the coordinates of the points are expressed as:

[0109]

[0110] Step 202. Bounding box generation

[0111] Analyze the relevant information for each sub-surface plate's 81 points:

[0112]

[0113] Create a bounding box based on the above range and name it: .

[0114] Step 203. Boundary line generation: generate the boundary line of the curved plate according to 81 points ( ):

[0115] ①

[0116] ②

[0117] ③

[0118] ④

[0119] Step 204. Plane plate topology analysis

[0120] For the "Ship—Block—PlanePanel—Boundary" node in the AM-XML file, extract the boundary points of the plane plate: extract the "StartPoint2d" and "Segment2d—Node2d" information from the "SimpleContour" subnode.

[0121] For a plane plate boundary point (x, y, z), determine whether it is within the bounding box:

[0122]

[0123] If the conditions are met, it is determined that there is a topological association between the plane plate and the sub-curved plate.

[0124] Step 205. Calculation of the base plane of the flat plate

[0125] For the topologically associated plane plate boundary point set , calculate the following parameters:

[0126]

[0127] Determine the minimum value , determine the base plane of the planar plate:

[0128] like Minimum, the base plane is YOZ;

[0129] like Minimum, the base plane is XOZ;

[0130] like Minimum, the base plane is XOY;

[0131] Step 3: Profile welding baseline NURBS curve reconstruction process. Specifically includes:

[0132] Based on profile welding baseline discrete point data , use the cumulative chord length method to calculate the non-uniform node vector. The specific steps are as follows:

[0133] Step 301. For a given point set , calculate the Euclidean distance between each pair of adjacent points: .

[0134] Step 302. Calculate the cumulative chord length, the cumulative chord length Defined as: ,final is the total chord length.

[0135] Step 303. Normalize the parameter value, normalize the cumulative chord length, and map it to the interval [0, 1]:

[0136]

[0137] Step 304. Construct a node vector. For a curve with an order of NURBS curve, node vector Constructed as: Repeat at the beginning and end times, that is The intermediate node values ​​are all parameter values , but remove the first and last ends.

[0138] Step 305. Obtained by the cumulative chord length method After constructing the parameter values, calculate the basis function matrix . Compute the B-spline basis functions using a recursive definition , construct the matrix :

[0139]

[0140] Step 306. Solve the linear equations according to the interpolation conditions. ,in is a column vector of control points, if For square ;if It is not a square matrix, and the least squares solution is required. .

[0141] Step 307. Set control point weights

[0142] Step 308. Generate profile welding baseline according to NURBS curve formula:

[0143]

[0144] According to the above NURBS curve definition and formula, the NURBS curve formula of each profile welding baseline can be calculated. Figure 2 The profile welding baseline information shown in . Save the NURBS curve information of each profile welding baseline on the surface, as well as the profile welding baseline discrete point information corresponding to the NURBS.

[0145] Step 4: Optimize the profile welding baseline NURBS curve, encrypt the NURBS curve discrete point set, and generate curve family III. Specifically include:

[0146] use The multiple profile welding baselines in step 3 are intercepted to obtain multiple new profile welding baselines.

[0147] Determine the "X", "Y", and "Z" values ​​of discrete points on the profile welding baseline and solve the corresponding parameters , is the difference between the maximum and minimum values ​​of "X", is the difference between the maximum and minimum values ​​of "Y", The difference between the maximum and minimum values ​​of the "Z" value. The minimum value in The minimum, then the profile welding baseline is parallel to the YOZ plane, that is, the base plane of the profile welding baseline is the YOZ plane; assuming The minimum, then the profile welding baseline is parallel to the XOZ plane, that is, the base plane of the profile welding baseline is the XOZ plane; assuming The minimum value is the profile welding baseline parallel to the XOY plane, that is, the base plane of the profile welding baseline is the XOY plane.

[0148] When the base plane of the profile welding baseline is XOY (the method for other directions is the same), fill the boundary line and Corresponding NURBS curves. For the discrete point data of the profile welding baseline and boundary line, they are arranged from small to large in terms of Z value, and then the curve family I of the curved plate is generated.

[0149] For multiple curves in the curve family I, the points are refined. Assuming the step size is λ=100mm, the "X" value of the point of the longitudinal section line refinement is: X=Xmin+λs and Xmin≤X≤Xmax, s=0,1,…. Each NURBS curve in the curve family I is known. When the X value is obtained, the points on each NURBS curve can be refined. That is, the formula of the known NURBS curve is converted, and the X value of a point on the curve is known, and the problem of the coordinates of the point can be solved.

[0150] Concatenate these encrypted points with the bounding box Make a judgment, assuming that the encrypted point is within the bounding box, then keep the encrypted point. The other points are deleted, and the encrypted points of the profile that exceed the curved plate boundary are deleted. Then the profile welding baseline NURBS curve is regenerated. and The corresponding NURBS curves of the corresponding curves are merged to form curve family II.

[0151] By connecting the corresponding points on multiple NURBS curves with the same X value, we can regenerate discrete points of curves perpendicular to multiple curves on curve family II, and reconstruct multiple curves of curve family III using the method in step 3. We can get curves in both the horizontal and vertical directions (curve family II and curve family III), and generate a grid-like segmented surface structure, such as Figure 3 shown.

[0152] Step 5, generating NURBS surface for the curved plate; specifically including:

[0153] After obtaining the data of curve family II and curve family III, since curve family II retains the original data characteristics compared to curve family III, and there is no error or deviation introduced by fitting, surface fitting is selected based on curve family II to ensure the geometric accuracy and consistency of the generated surface. First, solve the control point, weight and node vector information on each NURBS curve in curve family II.

[0154] Step 501. NURBS curve extraction and information acquisition:

[0155] For each NURBS curve in curve family II, its control points, weights and node vector information are extracted in turn to construct a complete NURBS parametric description.

[0156] Check the continuity and node distribution between curves, and record the necessary parameterized relationships for subsequent surface fitting optimization.

[0157] Step 502. Multi-curve fitting based on B-spline:

[0158] A B-spline mesh composed of curves is constructed to map the curve data of curve family II into the surface fitting framework.

[0159] Using the weighted average of the control points and the geometric constraints between the curves, a B-spline surface that conforms to the curve family II data is preliminarily generated, such as Figure 4 .

[0160] Constraints include continuity between curves (such as G0, G1 continuity) and smoothness of spatial distribution of control points to ensure the consistency of the surface in geometry and smoothness.

[0161] Step 503. Surface fitting optimization:

[0162] Use the least squares method or other fitting algorithms to optimize the surface and reduce the error between the fitted surface and the original curve family data.

[0163] Adjust control point positions, weight distribution, and node vector configuration to optimize the surface shape to make it closer to the geometric characteristics of the original data.

[0164] The error index is recalculated for each curve on the fitted surface to ensure that the deviations of all control points and specified sampling points are within the acceptable range.

[0165] Step 504. Dynamic adjustment and error evaluation:

[0166] After the surface is generated, the control points, weights, and node vector distribution are further adjusted through iterative optimization.

[0167] Combined with error evaluation methods (such as mean square error and maximum error), the fitting process is dynamically optimized to ensure that the quality of the generated surface meets the design requirements.

[0168] Special attention is paid to the fitting accuracy of boundary curves and the overall surface smoothness to avoid local distortion caused by boundary effects.

[0169] Step 505. Surface data storage and model reconstruction:

[0170] The final optimized B-spline surface parametric description (control points, weights, node vectors, surface equations) is stored in the database to form a standardized data record.

[0171] In summary, the three-dimensional design software for shipbuilding forming components and the curved surface forming detection interface method provided by the embodiment of the present invention extracts the curved plate data, flat plate data, and curved plate profile data of the AM-XML ship segment model, reconstructs the NURBS curve using the profile welding baseline discrete points on the curved plate, supplements two NURBS curves through the discrete boundaries of the curved plate, and intercepts the profile welding baseline NURBS curve according to the bounding box. Encrypt the points on the NURBS curve, generate the NURBS curve twice by corresponding to multiple points, and then generate the surface mesh of the curved plate, and finally regenerate the curved plate through multiple NURBS curves.

[0172] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A new type of shipbuilding component 3D design software and surface forming detection interface method, characterized in that: The method comprises the following steps: S1: Method for structured data parsing and database creation: extract data from AM-XML files and create a database. Use the segment name as the database name and the curved plate name as the sub-database name to achieve structured data parsing and database creation. S2: Data parsing and topological analysis method of multi-sub curved plate: Analyze the curved plate data, extract the boundary points and internal points of the curved plate, construct the bounding box of the curved plate, and use the bounding box to judge the boundary point data of the plane plate and analyze the topological relationship of the curved plate; S3: Profile welding baseline NURBS curve reconstruction process: Reconstruct the NURBS curve for the profile welding baseline discrete point data and save it in the database file; S4: Optimize the profile welding baseline NURBS curve, encrypt the discrete point set of the NURBS curve, and generate the curve family III; The profile welding baseline NURBS curve is intercepted by using a bounding box and encrypted to generate an encrypted point set. A new curve family III perpendicular to the profile welding baseline NURBS curve is generated by analyzing the encrypted point set; thus, the curved plate presents a grid-like distribution. S5: Generate NURBS surface for curved plate: Reconstruct the surface using the intercepted profile welding baseline NURBS curve.

2. A novel three-dimensional design software and surface forming detection interface method for shipbuilding forming components according to claim 1, characterized in that: The steps of solving the multi-sub-surface plate data parsing and topology analysis method in step 2 include: The boundary point information in the curved plate data is identified and judged to determine the boundary point information used for NURBS surface generation; and the boundary point information in the plane plate data is topologically analyzed to determine the connected plane plates.

3. A novel three-dimensional design software and curved surface forming detection interface method for shipbuilding forming components according to claim 2, characterized in that: The step S2 specifically includes the following sub-steps: S2.1 Data extraction and storage Further analyzing multiple sub-curved plate materials for a certain curved plate material corresponding to the profile welding baseline in step S1; and extracting node data, each sub-curved plate material contains k curves, and each curve corresponds to g spatial points; Save the point data of each sub-surface plate, and the coordinates of the points are expressed as: ； S2.2 generates a bounding box and parses the relevant information for each point of the sub-surface plate: ； Create a bounding box based on the above range and name it: ; S2.3 Generate boundary lines, generate boundary lines of curved plates based on point data : ； ； ； ； S2.4 Topological analysis of planar plates For the nodes in the AM-XML file, extract the boundary points of the plane plate; for a plane plate boundary point (x, y, z), determine whether it is within the bounding box: ； If the conditions are met, it is determined that the plane plate and the sub-curved plate have a topological association; S2.5 Calculation of base plane of planar plate For the topologically associated plane plate boundary point set , calculate the following parameters: ； Determine the minimum value , determine the base plane of the planar plate: like Minimum, the base plane is YOZ; like Minimum, the base plane is XOZ; like Minimum, base plane is XOY.

4. A novel three-dimensional design software and curved surface forming detection interface method for shipbuilding forming components according to claim 1, characterized in that: The step S3 specifically includes the following sub-steps: Based on profile welding baseline discrete point data , using the cumulative chord length method to calculate non-uniform node vectors; S3.1 For a given point set , calculate the Euclidean distance between each pair of adjacent points: ; S3.2 Calculate the cumulative chord length, cumulative chord length Defined as: ,final is the total chord length; S3.3 Normalize the parameter value, normalize the cumulative chord length, and map it to the interval [0, 1]: ； S3.4 constructs the node vector. For a NURBS curve with a curve order of p, the node vector Constructed as: Repeat at the beginning and end times, that is The intermediate node values ​​are all parameter values , but remove the first and last ends; S3.5 Obtained by the cumulative chord length method After constructing the parameter values, calculate the basis function matrix ; Compute B-spline basis functions using recursive definition , construct the matrix : ； S3.6 Solve the linear equations. Solve the linear equations according to the interpolation conditions. ,in is a column vector of control points, if For square ;if It is not a square matrix, and the least squares solution is required. ; S3.7 generates the profile welding baseline according to the NURBS curve formula: ； According to the above NURBS curve definition and formula, the NURBS curve formula of each profile welding baseline is calculated.

5. A novel three-dimensional design software and curved surface forming detection interface method for shipbuilding forming components according to claim 1, characterized in that: The specific process of step S4 is as follows: Using bounding box Intercepting the multiple profile welding baselines in step S3 to obtain multiple new profile welding baselines; Determine the "X", "Y", and "Z" values ​​of discrete points on the profile welding baseline and solve the corresponding parameters , is the difference between the maximum and minimum values ​​of "X", is the difference between the maximum and minimum values ​​of "Y", is the difference between the maximum and minimum values ​​of the "Z" value; The minimum value in The minimum, then the profile welding baseline is parallel to the YOZ plane, that is, the base plane of the profile welding baseline is the YOZ plane; assuming The minimum, then the profile welding baseline is parallel to the XOZ plane, that is, the base plane of the profile welding baseline is the XOZ plane; assuming The minimum, then the profile welding baseline is parallel to the XOY plane, that is, the base plane of the profile welding baseline is the XOY plane; Combined with the base plane of the profile welding baseline, the NURBS curve corresponding to the boundary line is filled; the discrete point data of the profile welding baseline and the boundary line are arranged from small to large in Z value, and then the curve family I of the curved plate is generated; For multiple curves in the curve family I, the points are refined. Assuming the step size is λ, the "X" value of the point where the longitudinal section line is refined is: X=Xmin+λs and Xmin≤X≤Xmax, s=0,1,…; each NURBS curve in the curve family I is known. When the X value is obtained, the points on each NURBS curve can be refined; that is, the formula of the known NURBS curve is converted, and the X value of a point on the curve is known, the coordinate problem of the point can be solved; The encrypted points and bounding box Make a judgment, assuming that the encrypted point is within the bounding box, then keep the encrypted point. The rest are deleted, and then the encrypted points of the profile that exceed the boundary of the curved plate are deleted; then the profile welding baseline NURBS curve is regenerated; and the NURBS curve corresponding to the curve corresponding to the boundary line is merged to form curve family II; By connecting corresponding points on multiple NURBS curves with the same X value, discrete curve points perpendicular to multiple curves on curve family II are regenerated, and multiple curves of curve family III are reconstructed using the method in step S3.

6. A novel three-dimensional design software and curved surface forming detection interface method for shipbuilding forming components according to claim 1, characterized in that: The sub-steps of step S5 are as follows: S5.1 NURBS curve extraction and information acquisition: For each NURBS curve in curve family II, extract its control points, weights and node vector information in turn to construct a complete NURBS parametric description; S5.2 Multi-curve fitting based on B-spline: Construct a B-spline mesh composed of curves and map the curve data of curve family II into the surface fitting framework; Using the weighted average of the control points and the geometric constraints between the curves, a B-spline surface that conforms to the curve family II data is preliminarily generated; the constraints include the continuity between the curves and the smoothness of the spatial distribution of the control points; S5.3 Surface fitting optimization: Use fitting algorithms to optimize the surface and reduce the error between the fitted surface and the original curve family data; adjust the control point position, weight distribution and node vector configuration to optimize the surface shape and make it closer to the geometric characteristics of the original data; Recalculate the error index for each curve on the fitting surface to ensure that the deviations of all control points and specified sampling points are within the threshold range; S5.4 Dynamic adjustment and error evaluation: After the surface is generated, the control points, weights and node vector distribution are further adjusted through iterative optimization; combined with the error evaluation method, the fitting process is dynamically optimized; S5.5 Surface data storage and model reconstruction: The final optimized B-spline surface parameter description: control points, weights, node vectors, surface equations are stored in the database to form standardized data records.

Images