Ship insulating material automatic laying system and method

Abstract

The invention relates to the technical field of ship design and manufacturing, in particular to a ship insulating material automatic laying system, which comprises a point cloud preprocessing module, a structure intelligent identification module, an insulating material adaptive modeling module and a collision detection optimization module, and is characterized in that the point cloud preprocessing module is used for preprocessing point cloud data of a ship structure; the structure intelligent identification module is used for analyzing geometric topological characteristics of the point cloud and automatically identifying typical structure types such as flat-bulb steel, T-shaped profiles and panels; the insulating material adaptive modeling module is used for adaptively generating an insulating material three-dimensional model matched with the structure according to the identified structure type, size and shape; and the collision detection optimization module is used for detecting geometric interference between an insulating material and a ship body structure and realizing interference-free assembly through adaptive adjustment, and has the advantages of higher intelligence, high adaptability, high efficiency, accuracy and the like.

Description

Technical Field

[0001] The invention relates to the technical field of ship design and manufacturing, in particular to an automatic laying system and method for ship insulation materials. Background Art

[0002] As we all know, laying of ship insulation materials is an important task in the process of ship design and construction, and is crucial to ensuring the safety, comfort and energy efficiency of ships. Traditionally, laying of insulation materials mainly relies on manual operation by designers, which has problems such as low efficiency, easy errors and poor adaptability. In recent years, with the increasing complexity of ship structures, manual laying methods have been unable to meet the design requirements of the modern shipbuilding industry.

[0003] To solve the above problems, the industry has proposed some rule-based or template-based auxiliary design methods for laying insulation materials. However, these methods still have the following shortcomings: (1) The identification of hull structure types is not intelligent enough and requires the pre-definition of a large number of rules or templates.

[0004] (2) The insulation material model cannot be adaptively adjusted according to the structure size and shape, resulting in unsatisfactory laying results.

[0005] (3) The degree of automation is not high, and a lot of manual intervention and post-processing are still required. Summary of the invention

[0006] The purpose of the present invention is to solve the above-mentioned deficiencies in the prior art and to provide a more intelligent, highly adaptable, efficient and accurate system and method for automatically laying ship insulation materials.

[0007] The technical solution adopted by the present invention to solve its technical problem is: A system for automatically laying insulating materials for ships, characterized in that it includes: a point cloud preprocessing module, a structure intelligent recognition module, an insulating material adaptive modeling module, and a collision detection optimization module. The point cloud preprocessing module, the structure intelligent recognition module, the insulating material adaptive modeling module, and the collision detection optimization module are interconnected. The point cloud preprocessing module is used to preprocess the point cloud data of the hull structure; the structure intelligent recognition module is used to analyze the geometric topological characteristics of the point cloud and automatically identify typical structural types such as bulb flat steel, T-profiles, and panels; the insulating material adaptive modeling module is used to adaptively generate a three-dimensional model of the insulating material that matches the structure according to the identified structural type, size, and shape; the collision detection optimization module is used to detect the geometric interference between the insulating material and the hull structure, and achieve interference-free assembly through adaptive adjustment.

[0008] A method for automatically laying ship insulation materials, characterized in that the method steps are as follows: Step S1: Analyze the geometric topological features of the point cloud and automatically identify three typical structural types: bulb flat steel, T-profile, and panel: Step S2: Based on the identified structural type, size and shape parameters, a three-dimensional model of the insulating material matching the structure is adaptively generated.

[0009] The specific steps of step S1 of the present invention are as follows: (1) Extract the coordinates (x, y, z) of all element points to form a coordinate array pointsArray; (2) Calculate the parameters (A, B, C, D) of the best fitting plane using the least squares method or other fitting algorithms; (3) Traverse each element point P (x0, y0, z0) in the coordinate array pointsArray and calculate its distance d to the fitting plane: d = |Ax0 + By0 + Cz0 + D| / sqrt(A^2 + B^2 + C^2) Compare the distance d with the preset threshold threshold. If d <= threshold, mark it as inlier, otherwise mark it as outlier. (4) Count the marking status of all element points: If the vast majority of points (e.g., more than 90%) are marked as inliers, then these points are considered to be basically in the same plane; If a considerable number of points (e.g., more than 10%) are marked as outliers, these points are considered not to be in the same plane; (5) Judging from the statistical results: If the points are basically in the same plane, we further determine whether their normal vectors are consistent to confirm whether it is a panel. If the points are not in the same plane, we can rule out the possibility of a panel and further analyze other structural types. (6) For the suspected spherical flat steel structure, extract its three-dimensional coordinate point cloud data, select m equally spaced cross-sectional positions along the length direction, and form a set S; (7) For each cross-sectional position sj, extract the point set Qj located in the cross-sectional area, calculate its geometric center cj and the principal axis directions v1 and v2, as well as the lengths l1j and l2j on the principal axis; (8) Determine the cross-sectional characteristics of the bulb flat steel: Whether the length ratio of the main axis of the section is within the preset range, i.e., l1j / l2j∈[r1, r2]; Whether the distance between the geometric centers of adjacent sections is within the preset range, i.e., dj, j+1∈[d1, d2]; Whether the direction of the main axis of the section changes smoothly, and whether the angles between the main axes of adjacent sections are within the preset range; If all sections meet the preset conditions, it is judged to be a bulb-flat steel structure; (9) For the suspected T-profile structure, extract its 3D point cloud data and divide it into n cross sections; (10) For each cross section, fit it into two mutually perpendicular rectangles through a fitting algorithm, and extract the geometric characteristic parameters such as rectangle width wi, height hi, centroid coordinates (xci, yci, zci), moments of inertia Ixi and Iyi, and shape factors Sxi and Syi; (11) Determine the cross-sectional characteristics of the T-profile: Whether the width and height of the rectangle meet the size ratio relationship, that is, hi>wi, twi <wi,thi<hi; Whether the centroid is located on the centerline of the web and below the upper surface of the flange; Whether the matching degree p between the cross-section parameters and the standard value of the T-profile is greater than the preset threshold; Whether adjacent sections are parallel and equidistant, i.e., di≈di+1, θi≈0; If most of the cross sections meet the above conditions, it is judged to be a T-profile structure; The specific steps of step S2 of the present invention are as follows: (1) For the bulb-flat steel structure, an insulating layer entity with uniform thickness is adaptively generated on its surface according to the cross-sectional position and size parameters; (2) For T-profile structures, a matching parametric model of the insulation layer is adaptively generated based on the key dimensions of the flange and web, and the insulation layer entity is instantiated; (3) For the panel structure, an insulating layer entity with uniform thickness is generated on its surface according to the fitting plane equation and boundary contour.

[0010] Compared with the prior art, the present invention has the following beneficial effects: (1) By using point cloud data as input and intelligently analyzing the geometric topological features of the point cloud, the typical structural types in the hull can be identified efficiently and accurately without the need to predefine identification rules and templates, which greatly improves the intelligence level of insulation material laying.

[0011] (2) For different structural types, a matching parametric three-dimensional model of the insulation material is adaptively generated, which can be automatically adjusted according to the actual structure size and shape to generate an insulation layer that is highly consistent with the hull structure, achieving an ideal laying effect.

[0012] (3) The fully automated design process minimizes manual operations and errors, and improves design efficiency by several times compared to manual methods, significantly shortening the shipbuilding cycle and saving costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 The coordinate position relationship diagram of each element point P.

[0014] Figure 2 This is the analysis diagram for the oblate spherical steel structure.

[0015] Figure 3 This is the judgment and analysis diagram of the T-profile structure.

[0016] Figure 4 This is a coordinate dynamic change diagram of Example 1.

[0017] Figure 5 This is a coordinate dynamic change diagram of Example 2. DETAILED DESCRIPTION

[0018] The present invention is further described below in conjunction with the accompanying drawings: As shown in the accompanying drawings, a system for automatically laying insulating materials for ships is characterized in that it includes: a point cloud preprocessing module, a structure intelligent recognition module, an insulating material adaptive modeling module, and a collision detection optimization module. The point cloud preprocessing module, the structure intelligent recognition module, the insulating material adaptive modeling module, and the collision detection optimization module are interconnected. The point cloud preprocessing module is used to preprocess the point cloud data of the hull structure; the structure intelligent recognition module is used to analyze the geometric topological characteristics of the point cloud, and automatically identify typical structural types such as bulb flat steel, T-profiles, and panels; the insulating material adaptive modeling module is used to adaptively generate a three-dimensional model of the insulating material that matches the structure according to the identified structural type, size, and shape; the collision detection optimization module is used to detect the geometric interference between the insulating material and the hull structure, and achieve interference-free assembly through adaptive adjustment.

[0019] A method for automatically laying ship insulation materials, characterized in that the method steps are as follows: Step S1: Analyze the geometric topological features of the point cloud and automatically identify three typical structural types: bulb flat steel, T-profile, and panel: Step S2: Based on the identified structural type, size and shape parameters, a three-dimensional model of the insulating material matching the structure is adaptively generated.

[0020] The specific steps of step S1 of the present invention are as follows: (1) Extract the coordinates (x, y, z) of all element points to form a coordinate array pointsArray; (2) Calculate the parameters (A, B, C, D) of the best fitting plane using the least squares method or other fitting algorithms; (3) Traverse each element point P (x0, y0, z0) in the coordinate array pointsArray and calculate its distance d to the fitting plane: d = |Ax0 + By0 + Cz0 + D| / sqrt(A^2 + B^2 + C^2) Compare the distance d with the preset threshold threshold. If d <= threshold, mark it as inlier, otherwise mark it as outlier. (4) Count the marking status of all element points: If the vast majority of points (e.g., more than 90%) are marked as inliers, then these points are considered to be basically in the same plane; If a considerable number of points (e.g., more than 10%) are marked as outliers, these points are considered not to be in the same plane; (5) Judging from the statistical results: If the points are basically in the same plane, we further determine whether their normal vectors are consistent to confirm whether it is a panel; if the points are not in the same plane, we can rule out the possibility of a panel and further analyze other structural types ( Figure 1 The coordinate position relationship diagram of each element point P is shown); (6) For the suspected spherical flat steel structure, extract its three-dimensional coordinate point cloud data, select m equally spaced cross-sectional positions along the length direction, and form a set S; (7) For each cross-sectional position sj, extract the point set Qj located in the cross-sectional area, calculate its geometric center cj and the principal axis directions v1 and v2, as well as the lengths l1j and l2j on the principal axis; (8) Determine the cross-sectional characteristics of the bulb flat steel: Whether the length ratio of the main axis of the section is within the preset range, i.e., l1j / l2j∈[r1, r2]; Whether the distance between the geometric centers of adjacent sections is within the preset range, i.e., dj, j+1∈[d1, d2]; Whether the direction of the section principal axis changes smoothly, and whether the angles between the adjacent section principal axes are within the preset range ( Figure 2 It is the judgment and analysis diagram of the oblate spherical steel structure); If all sections meet the preset conditions, it is judged to be a bulb-flat steel structure; (9) For the suspected T-profile structure, extract its 3D point cloud data and divide it into n cross sections; (10) For each cross section, fit it into two mutually perpendicular rectangles through a fitting algorithm, and extract the geometric characteristic parameters such as rectangle width wi, height hi, centroid coordinates (xci, yci, zci), moments of inertia Ixi and Iyi, and shape factors Sxi and Syi; (11) Determine the cross-sectional characteristics of the T-profile: Whether the width and height of the rectangle meet the size ratio relationship, that is, hi>wi, twi <wi,thi<hi; Whether the centroid is located on the centerline of the web and below the upper surface of the flange; Whether the matching degree p between the cross-section parameters and the standard value of the T-profile is greater than the preset threshold; Whether adjacent sections are parallel and equidistant, i.e., di≈di+1, θi≈0; If most of the sections meet the above conditions, they are judged to be T-shaped structures ( Figure 3 is the judgment and analysis diagram of the T-profile structure); The specific steps of step S2 of the present invention are as follows: (1) For the bulb-flat steel structure, an insulating layer entity with uniform thickness is adaptively generated on its surface according to the cross-sectional position and size parameters; (2) For T-profile structures, a matching parametric model of the insulation layer is adaptively generated based on the key dimensions of the flange and web, and the insulation layer entity is instantiated; (3) For the panel structure, an insulating layer entity with uniform thickness is generated on its surface according to the fitting plane equation and boundary contour.

[0021] Compared with the prior art, the present invention has the following beneficial effects: (1) By using point cloud data as input and intelligently analyzing the geometric topological features of the point cloud, the typical structural types in the hull can be identified efficiently and accurately without the need to predefine identification rules and templates, which greatly improves the intelligence level of insulation material laying.

[0022] (2) For different structural types, a matching parametric three-dimensional model of the insulation material is adaptively generated, which can be automatically adjusted according to the actual structure size and shape to generate an insulation layer that is highly consistent with the hull structure, achieving an ideal laying effect.

[0023] (3) The fully automated design process minimizes manual operations and errors, and improves design efficiency by several times compared to manual methods, significantly shortening the shipbuilding cycle and saving costs.

[0024] 1. Calculating the distance from the element point to the fitting plane is a key step in determining whether the points are in the same plane. Assume that an optimal plane has been obtained through some fitting method (such as least squares method), and its equation can be expressed as: Ax + By + Cz + D = 0 Where (A, B, C) is the unit normal vector of the plane, (x, y, z) is the coordinates of any point on the plane, and D is the constant term of the plane equation.

[0025] For any element point P (x0, y0, z0) on the structure to be judged, its distance d to the fitting plane can be calculated by the following formula: d = |Ax0 + By0 + Cz0 + D| / sqrt(A^2 + B^2 + C^2) Here, |·| represents the absolute value, and sqrt(·) represents the square root. The denominator sqrt(A^2 + B^2 + C^2) is used to normalize the plane normal vector to ensure that the distance calculation result is independent of the coordinate system.

[0026] You can follow the steps below: 1. Extract the coordinates (x, y, z) of all element points to form a coordinate array pointsArray.

[0027] 2. Based on the coordinate array pointsArray, use the least squares method or other fitting algorithm to calculate the parameters (A, B, C, D) of the best fitting plane.

[0028] 3. Traverse each element point P(x0, y0, z0) in the coordinate array pointsArray: (1) According to the above distance formula, calculate the distance d from P to the fitting plane; (2) Compare the distance d with the preset threshold: - If d <= threshold, the point is considered to be in the plane and marked as inlier; - If d>threshold, the point is considered to deviate from the plane and marked as outlier.

[0029] 4. Count the markings of all element points: - If the vast majority of points (e.g. more than 90%) are marked as inliers, then these points are considered to be basically in the same plane; - If a considerable number of points (e.g. more than 10%) are marked as outliers, these points are considered not to be in the same plane.

[0030] 5. Make the final judgment based on the statistical results: - If the points are basically in the same plane, we further determine whether their normal vectors are consistent to confirm whether it is a panel; - If the points are not in the same plane, the possibility of panels can be ruled out and other structural types can be further analyzed.

[0031] Second, the method of judging the ball flat steel, Assume that the three-dimensional coordinate point cloud data of the bulb flat steel surface has been obtained, recorded as the set $P=\{p_i=(x_i,y_i,z_i)|i=1,2,\dots,n\}$, where $n$ is the number of points. It is necessary to determine whether these points meet the characteristics of the bulb flat steel.

[0032] 1. Select the section location: First, we need to select some cross-sectional positions, which are recorded as a set $S=\{s_j|j=1,2,\dots,m\}$, where $m$ is the number of cross sections. These cross-sectional positions can be selected at equal intervals along the length of the bulb flat steel.

[0033] 2. Extract points on the cross section: For each section position $s_j$, extract the points located on the section from the point cloud data $P$, recorded as the set $Q_j=\{p_i=(x_i,y_i,z_i)|z_i=s_j,p_i\in P\}$.

[0034] 3. Calculate the geometric center of the cross section: For each section $Q_j$, calculate its geometric center $c_j=(x_{cj},y_{cj},z_{cj})$, where: $x_{cj}=\frac{1}{|Q_j|}\sum_{p_i\in Q_j}x_i$ $y_{cj}=\frac{1}{|Q_j|}\sum_{p_i\in Q_j}y_i$ $z_{cj}=s_j$ 4. Calculate the principal axis direction and length of the cross section: For each section $Q_j$, calculate its principal axis direction and length. First, translate the points on the section to the geometric center to obtain a new point set $Q_j'=\{p_i'=p_i-c_j|p_i\in Q_j\}$. Then, perform singular value decomposition (SVD) on $Q_j'$ to obtain the left singular vector matrix $U$, the singular value matrix $\Sigma$, and the right singular vector matrix $V$. Among them, the first two column vectors $v_1$ and $v_2$ of the right singular vector matrix $V$ represent the two principal axis directions of the section respectively. The length of the section in these two principal axis directions can be calculated: $l_{1j}=\max_{p_i\in Q_j}(p_i\cdot v_1)-\min_{p_i\in Q_j}(p_i\cdot v_1)$ $l_{2j}=\max_{p_i\in Q_j}(p_i\cdot v_2)-\min_{p_i\in Q_j}(p_i\cdot v_2)$ 5. Determine whether it meets the characteristics of bulb flat steel: According to the design requirements of the bulb flat steel, some judgment conditions can be set, such as: - Whether the ratio of the principal axis lengths of the cross section is within a certain range, that is, $\frac{l_{1j}}{l_{2j}}\in[r_1,r_2]$, where $r_1$ and $r_2$ are the preset ratio ranges.

[0035] - Whether the geometric center distance between adjacent sections is within a certain range, that is, $d_{j,j+1}=\sqrt{(x_{cj}-x_{c,j+1})^2+(y_{cj}-y_{c,j+1})^2+(z_{cj}-z_{c,j+1})^2}\in[d_1,d_2]$, where $d_1$ and $d_2$ are the preset distance ranges.

[0036] - Whether the change of the main axis direction of the section is smooth, the angle between the main axis directions of adjacent sections can be calculated to determine whether it is within a certain range.

[0037] If all sections meet the preset judgment conditions, it can be considered that the point cloud data meets the characteristics of bulb flat steel.

[0038] Through the above steps, the three-dimensional coordinate point cloud data of the bulb flat steel can be analyzed to determine whether it meets the preset feature requirements.

[0039] 3. The following is a method for identifying T profiles: The method first obtains the three-dimensional point cloud data of the T-profile and divides it into multiple sections. Then, each section is analyzed and its geometric characteristic parameters are calculated, including the section shape, size, centroid position, moment of inertia and shape factor. Then, these parameters are compared with the standard values ​​of the T-profile to determine whether each section meets the geometric characteristics of the T-profile. Finally, based on the section analysis results, a comprehensive judgment is made as to whether the model is a T-profile. The present invention provides a fast, accurate and automated T-profile recognition method, which can be widely used in industrial design, quality inspection and other fields. Specific implementation method: 1. Obtain the three-dimensional point cloud data of the T-shaped profile and divide it into multiple cross-sections. Suppose there are n cross-sections in total, and each cross-section contains m points. The coordinates of the j-th point in the i-th cross-section are $(x_{ij}, y_{ij}, z_{ij})$.

[0041] 2. For each cross-section, extract its geometric feature parameters: (1) Cross-section shape: Through a fitting algorithm, fit the cross-section into two mutually perpendicular rectangles. Suppose the widths and heights of the fitted rectangles are $w_i$ and $h_i$ respectively.

[0042] (2) Cross-section dimensions: Calculate the width $w_i$ and height $h_i$ of the rectangle, and determine whether they satisfy the dimensional ratio relationship of the T-shaped profile, that is, $h_i>w_i$, and $t_{wi}<w_i$, $t_{hi}<h_i$, where $t_{wi}$ and $t_{hi}$ are the thicknesses of the web and flange respectively.

[0043] (3) Cross-section centroid: Calculate the centroid coordinates $(x_{ci}, y_{ci}, z_{ci})$ of the cross-section, where $x_{ci}= \frac{1}{m}\sum_{j=1}^m x_{ij}$, $y_{ci} = \frac{1}{m}\sum_{j=1}^m y_{ij}$, $z_{ci} = \frac{1}{m}\sum_{j=1}^m z_{ij}$. Determine whether the centroid is located on the center line of the web and below the upper surface of the flange.

[0044] (4) Cross-section moment of inertia: According to the dimensions of the fitted rectangle, calculate the moments of inertia $I_{xi}$ and $I_{yi}$ about the horizontal and vertical axes.

[0045] - $I_{xi} = \frac{b_{fi} h_{fi}^3}{12} + \frac{b_{wi} (h_i-h_{fi})^3}{12}$ - $I_{yi} = \frac{h_{fi} b_{fi}^3}{12} + \frac{(h_i-h_{fi}) b_{wi}^3}{12}$ (5) Cross-section shape factor: According to the moment of inertia and cross-section dimensions, calculate the shape factors $S_{xi}$ and $S_{yi}$ about the horizontal and vertical axes.

[0046] - $S_{xi} = \frac{I_{xi}}{h_i / 2}$ - $S_{yi} = \frac{I_{yi}}{b_{fi} / 2}$ 3. Compare the geometric characteristic parameters of each section with the standard value of the T profile to determine whether it meets the requirements of the T profile. Assume that the number of sections that meet the requirements is k, then the matching degree of the T profile is $p = \frac{k}{n}$.

[0047] 4. Calculate the distance and angle between adjacent sections to determine whether they are parallel and equidistant. Let the distance between adjacent sections be $d_i$ and the angle be $\theta_i$, then $d_i \approx d_{i+1}$ and $\theta_i \approx 0$ should be satisfied.

[0048] 5. Based on the cross-sectional analysis results and cross-sectional distribution, determine whether the model is a T-shaped profile. If the matching degree $p$ is greater than the threshold and the adjacent cross-sections are parallel and equidistantly distributed, the model is considered to be a T-shaped profile, otherwise it is considered not to be a T-shaped profile.

[0049] 6. Display the recognition results in the form of text or graphics, and annotate the key geometric parameters and matching degree so that users can intuitively understand the characteristics of the model.

[0050] The T-profile identification method of the present invention can quickly and accurately determine whether a model is a T-profile by analyzing the cross-sectional geometric features of three-dimensional point cloud data.

[0051] Embodiment 1: Automatic laying method of ship bulb flat steel insulation material: Assume that the width of the insulating material (from A to F) is a fixed value L, and the thickness is a fixed value T. The positions of the dimensional points B, C, D, and E of the bulb flat steel can change, resulting in dynamic changes in the coordinates of the corresponding points ( Figure 4 The following is a detailed analysis of the calculation method of each point coordinate.

[0052] Define the following variables and parameters: Known conditions: L: The fixed width of the insulating material, that is, the distance from A to F.

[0053] T: Fixed thickness of insulating material.

[0054] B(x_B, y_B): coordinates of the fixed point B of the bulb flat.

[0055] H: The height of the bulb flat, that is, the distance from B to C.

[0056] W: The width of the ball head, that is, the distance from C to D.

[0057] According to the known conditions, the coordinates of each point are calculated as follows: Coordinate calculation of point A: x_A = x_B - (L - W - H) / 2 y_A = y_B Coordinates of point B: x_B (known) y_B (known) Calculation of coordinates of point C: x_C = x_B y_C = y_B + H Calculation of coordinates of point D: x_D = x_B + W y_D = y_B + H Calculation of coordinates of point E: x_E = x_B + W y_E = y_B Calculation of coordinates of point F: x_F = (x_B + W + (L - W - 2 * H)) / 2 y_F = y_B Calculation of coordinates of point A': x_A' ​​= x_A y_A' = y_A + T Calculation of coordinates of point B': x_B' = x_B - T y_B' = y_B + T Calculation of coordinates of point C': x_C' = x_C - T y_C' = y_C + T Calculation of coordinates of point D': x_D' = x_D + T y_D' = y_D + T Calculation of coordinates of point E': x_E' = x_E + T y_E' = y_E + T Calculation of coordinates of point F': x_F' = x_F y_F' = y_F + T Through the above calculations, the coordinates of the insulating material and the bulb flat steel structure are obtained. Among them, the coordinates of point A and point F ensure that the length of the insulating material on the left and right sides of the bulb flat steel meets the design requirements, the coordinates of point C and point D reflect the height and width of the bulb flat steel, and the coordinates of points A', B', C', D', E' and F' take into account the thickness of the insulating material based on the original points.

[0058] When the size of the bulb flat steel changes, the adaptive modeling and laying of the insulation material and the bulb flat steel structure can be achieved by simply recalculating the coordinates of each point according to the new bulb flat steel height H and width W. This parametric modeling method greatly improves the efficiency of insulation material design and assembly, while ensuring the optimization of insulation performance and structural bulb flat steel matching.

[0059] Through the above steps, the automatic modeling and laying of insulation materials based on the dynamic size of the bulb flat steel is realized. This method uses parametric modeling and coordinate transformation technology to generate a matching insulation layer model in real time according to the size change of the bulb flat steel, and realizes the adaptive assembly of the insulation material and the bulb flat steel structure. Embodiment 2: Automatic laying method of T-profile insulation material for ships: This embodiment provides a modeling method for automatic laying of T-profile insulating materials for ships. The method obtains the positioning points and specifications of the T-profile, dynamically adjusts the size parameters of the T-profile using the changes in the point coordinates, adaptively generates a matching insulating material model, and realizes the automatic assembly of the insulating material and the T-profile.

[0060] Specifically, the modeling method of this embodiment includes the following steps: Assume that A' is the reference point with coordinates (0, 0), and then calculate the coordinates of other points based on the obtained T-profile size and insulation layer thickness. Define the main dimensions of the T-profile: h: web height (mm), b: flange width (mm), t: web thickness (mm), s: flange thickness (mm), T: insulation layer thickness (mm) Then the coordinates of the other points are respectively.

[0061] A': (0, 0) A: (0-T, 0) B: (0-T, hsT) B': (0, hs) C': (0-(bt) / 2, hs) C: (0-(bt) / 2-T, hsT) D: (0-(bt) / 2-T, h + T) D': (0-(bt) / 2, h) E': (b-(bt) / 2, h) E: (b-(bt) / 2 + T, h + s) F: (b-(bt) / 2 + T, hsT) F': (b-(bt) / 2, hs) G': (s, hs) G: (s+T, hsT) H: (s+T, 0) H': (s, 0) In this way, each coordinate point of the insulating material is calculated, thereby realizing the laying of the insulating material.

[0062] When the specifications of the T-profile change, the model of the insulating material can be adaptively adjusted by simply recalculating the coordinates of each point according to the new size parameters to ensure the best matching and assembly of the insulating material and the T-profile.

[0063] Generate an insulation layer model. Based on the adjusted T-profile coordinate point A', automatically generate a matching insulation layer model. The insulation layer thickness can be parameterized according to material properties and design requirements. Then use the stretching operation to create an insulation material solid model that matches the shape of the bulb flat steel.

[0064] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0065] Embodiment 3: Automatic laying method of ship bracket and panel insulation material: The present invention provides an automatic laying method of insulating materials based on hull structure feature recognition. The method extracts coordinate points of hull structures such as brackets and panels, determines the structure type, and lays insulating materials of a specified thickness in a direction perpendicular to the structural surface, thereby achieving automatic matching of the insulating layer and the hull structure.

[0066] Figure 2 The schematic diagram of automatically laying the insulating material of the ship bracket structure by the method of the present invention is shown. In the figure, A, B, C, D, E, F, and G represent the characteristic coordinate points of the bracket structure.

[0067] After extracting the characteristic coordinate points of the bracket ABCDEFG, the structure is determined to be a bracket according to the existence of points B, C, D, and E. Then, an insulating material with a parameterized thickness is generated in the direction perpendicular to the bracket surface. The contour of the insulating material is consistent with the bracket contour, realizing automatic matching between the two.

[0068] The insulation laying method of the present invention does not require manual operation, and only needs to extract the three-dimensional model of the hull structure to automatically complete the modeling and laying of the insulation material. The method realizes intelligent recognition of the hull structure by extracting coordinate points and judging the structure type, and realizes rapid laying of the insulation layer by generating insulation materials in the vertical direction, and has the advantages of high modeling efficiency, high laying accuracy, strong adaptability, etc.

[0069] The above are only preferred embodiments of the present invention. Those skilled in the art may make various improvements thereto according to design requirements, but the basic principles thereof should not exceed the scope of protection claimed by the present invention.

Claims

1. An automatic laying system for ship insulation materials, characterized in that: It includes: a point cloud preprocessing module, a structure intelligent recognition module, an insulation material adaptive modeling module, and a collision detection optimization module. The point cloud preprocessing module, the structure intelligent recognition module, the insulation material adaptive modeling module, and the collision detection optimization module are interconnected. The point cloud preprocessing module is used to preprocess the hull structure point cloud data; The intelligent structural recognition module is used to analyze the geometric topological features of the point cloud and automatically identify typical structural types such as bulb flat steel, T-profiles, and panels; The insulation material adaptive modeling module is used to adaptively generate a three-dimensional insulation material model that matches the structure based on the identified structure type, size and shape; The collision detection optimization module is used to detect the geometric interference between the insulation material and the hull structure, and achieve interference-free assembly through adaptive adjustment.

2. A method for automatically laying ship insulation materials, characterized in that The steps are as follows: Step S1: Analyze the geometric topological features of the point cloud and automatically identify three typical structural types: bulb flat steel, T-profile, and panel: Step S2: Based on the identified structural type, size and shape parameters, adaptively generate a three-dimensional model of the insulating material that matches the structure.

3. The method for automatically laying ship insulation materials according to claim 1, characterized in that The specific steps of step S1 are as follows: (1) Extract the coordinates (x, y, z) of all element points to form a coordinate array pointsArray; (2) Calculate the parameters (A, B, C, D) of the best fitting plane using the least squares method or other fitting algorithms; (3) Traverse each element point P (x0, y0, z0) in the coordinate array pointsArray and calculate its distance d to the fitting plane: d = |Ax0 + By0 + Cz0 + D| / sqrt(A^2 + B^2 + C^2) Compare the distance d with the preset threshold threshold. If d <= threshold, mark it as inlier, otherwise mark it as outlier. (4) Count the markings of all element points: If the vast majority of points (e.g., more than 90%) are marked as inliers, then these points are considered to be basically in the same plane; If a considerable number of points (e.g., more than 10%) are marked as outliers, these points are considered not to be in the same plane; (5) Judging from the statistical results: If the points are basically in the same plane, we further determine whether their normal vectors are consistent to confirm whether it is a panel; If the points are not in the same plane, the possibility of a panel can be ruled out and other structural types can be further analyzed; (6) For the suspected spherical flat steel structure, extract its three-dimensional coordinate point cloud data, select m equally spaced cross-sectional positions along the length direction, and form a set S; (7) For each cross-sectional position sj, extract the point set Qj located in the cross-sectional area, calculate its geometric center cj and the principal axis directions v1 and v2, as well as the lengths l1j and l2j on the principal axis; (8) Determine the cross-sectional characteristics of the bulb flat steel: Whether the length ratio of the main axis of the section is within the preset range, i.e., l1j / l2j∈[r1, r2]; Whether the distance between the geometric centers of adjacent sections is within the preset range, i.e., dj, j+1∈[d1, d2]; Whether the direction of the main axis of the section changes smoothly, and whether the angles between the main axes of adjacent sections are within the preset range; If all sections meet the preset conditions, it is judged to be a bulb-flat steel structure; (9) For the suspected T-profile structure, extract its 3D point cloud data and divide it into n cross sections; (10) For each cross section, fit it into two mutually perpendicular rectangles through a fitting algorithm, and extract the geometric characteristic parameters such as rectangle width wi, height hi, centroid coordinates (xci, yci, zci), moments of inertia Ixi and Iyi, and shape factors Sxi and Syi; (11) Determine the cross-sectional characteristics of the T-profile: Whether the width and height of the rectangle meet the size ratio relationship, that is, hi>wi, twi <wi,thi<hi; Whether the centroid is located on the centerline of the web and below the upper surface of the flange; Whether the matching degree p between the cross-section parameters and the standard value of the T-profile is greater than the preset threshold; Whether adjacent sections are parallel and equidistant, i.e., di≈di+1, θi≈0; If most of the cross sections meet the above conditions, it is judged to be a T-profile structure.

4. The method for automatically laying ship insulation materials according to claim 1, characterized in that The specific steps of step S2 are as follows: (1) For the bulb-flat steel structure, an insulating layer entity with uniform thickness is adaptively generated on its surface according to the cross-sectional position and size parameters; (2) For T-profile structures, a matching parametric model of the insulation layer is adaptively generated based on the key dimensions of the flange and web, and the insulation layer entity is instantiated; (3) For the panel structure, an insulating layer entity with uniform thickness is generated on its surface according to the fitting plane equation and boundary contour.

Images