Digital modeling method for building decoration project marking

Abstract

The invention relates to the technical field of digital construction of building engineering, and discloses a digital modeling method for building decoration engineering marking, and the method comprises the steps: reconstructing an original point cloud of a building base plane into a discrete manifold triangular mesh, and extracting a differential geometric attribute; calculating an algebraic sign distance based on the process tolerance, and constructing a global tolerance potential energy field; under the constraint of an anchor drift feasible region, solving an optimal mark path which minimizes a composite energy functional containing tension and tolerance potential energy; generating a semantic envelope body by taking the path as a center, calculating a Boolean interference relationship between the semantic envelope body and the grid, dividing an effective area, a polishing area and a repairing area, and calculating a processing depth; and finally, outputting the digital marking model and correcting the building information model. According to the method, the basic plane defects are automatically avoided through a potential energy field guiding path, digitization and quantification of the construction work amount are achieved through a semantic envelope body, closed-loop feedback of a building information model and a physical field geometric state is established, and the construction precision and efficiency are improved.

Description

Technical Field

[0001] This invention relates to the field of digital construction technology in building engineering, specifically a digital modeling method for marking building decoration projects. Background Technology

[0002] In modern architectural finishing projects, Building Information Modeling (BIM) technology has been widely used in the detailed design and construction simulation stages. Theoretical design models usually assume that the building base surface (such as concrete walls and floors) is an ideal geometric plane. However, due to the limitations of civil engineering construction technology, the actual delivered physical base surface inevitably has random geometric deviations, which manifest as global wavy undulations or local bulges and depressions.

[0003] Current on-site marking and layout operations typically use total stations or laser line projectors to directly project the theoretical design coordinates from the BIM model onto the physical base surface. This rigid projection method strictly adheres to the theoretical geometric straight lines, without considering the actual topological characteristics of the physical base surface. When the theoretical marking path passes through raised areas of the base surface, in order to meet the flatness acceptance standards for the decorative surface layer, the construction team is often forced to physically grind large areas of the structural base layer, forcing the base surface to conform to the theoretical lines. This method not only consumes a lot of labor and machinery costs and generates dust pollution, but may even damage the protective layer of the building structure's steel reinforcement due to excessive grinding. Conversely, when the path passes through deeply recessed areas, a large amount of leveling material is required for backfilling, increasing material costs and curing time.

[0004] Furthermore, traditional marking and acceptance processes lack digital and quantitative analysis methods for the interference relationship between base surface flatness and construction paths. Judgment of grinding or repair areas relies primarily on on-site personnel using straightedges or feeler gauges for discrete point sampling, failing to generate continuous, comprehensive quantitative data. Simultaneously, on-site positioning adjustments due to base surface defects often remain at the physical operation level; the actual data after adjustment cannot be automatically transmitted back to the BIM model, resulting in a separation between the digital model and the physical entity. This data gap means that subsequent prefabricated components (such as keels and decorative panels) are still produced based on original theoretical data, making dimensional deviations highly likely during installation and severely restricting the accuracy and efficiency of prefabricated decoration technology. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a digital modeling method for marking architectural decoration projects, which solves the problem that theoretically designed straight lines are difficult to adapt to the random geometric defects of the physical base surface during the traditional architectural decoration layout process.

[0006] To achieve the above objectives, the present invention provides the following technical solution: The first aspect of the present invention provides a digital modeling method for marking architectural decoration engineering projects. The method includes: firstly, acquiring original point cloud data of the building base surface and reconstructing it into a discrete manifold triangular mesh with topological continuity; subsequently, performing plane fitting on the discrete manifold triangular mesh to determine an ideal reference plane; calculating the algebraic sign distance of the mesh vertices relative to the ideal reference plane based on a preset process tolerance threshold, and constructing a global tolerance potential energy field accordingly; simultaneously, obtaining theoretical start and end points from a building information model, constructing a feasible region for anchor point drift based on the theoretical start and end points, and establishing a system including path tension and... The composite energy functional of external potential energy is solved under the constraint of the feasible region of anchor point drift to find the optimal marking path that minimizes the composite energy functional. Then, the optimal marking path is used as the central axis to generate a semantic envelope. The Boolean interference relationship between the semantic envelope and the discrete manifold triangular mesh is calculated. Based on the Boolean interference relationship and the directional attribute of the algebraic symbol distance, the region is divided and the construction processing depth is calculated. Finally, based on the results of the region division and the construction processing depth, a digital marking model containing construction attributes is generated to guide on-site operations. At the same time, the spatial coordinates of the optimal marking path are used to correct the component positioning data corresponding to the building decoration engineering markings in the building information model.

[0007] In one optional implementation, to improve the accuracy and topology quality of the base surface reconstruction, the process of acquiring the original point cloud data of the building base surface and reconstructing it into a discrete manifold triangular mesh specifically performs the following processing: Statistical denoising is performed on the acquired original point cloud data, and outlier noise points exceeding the distance threshold are removed by calculating the average Euclidean distance between the target point and its nearest neighbors; voxel grid downsampling is performed on the denoised point cloud data, using the geometric centroid of the point cloud data within the voxel grid as a representative point to achieve data homogenization; the normal vector of the downsampled point cloud data is calculated, and the minimum spanning tree algorithm is used to unify the direction of the normal vector, resulting in a clean point cloud dataset with normal vectors.

[0008] Furthermore, based on the clean point cloud dataset, an implicit surface function is constructed using the Poisson surface reconstruction algorithm. The indicator function is obtained by solving the Poisson equation, and the isosurface is extracted to generate an initial triangular mesh. Subsequently, a topology check is performed on the initial triangular mesh to detect and remove non-manifold edges shared by more than two triangles and non-manifold vertices that cannot form a single connected sector, thereby obtaining a triangular mesh that satisfies the manifold property, providing a geometric basis for subsequent Boolean operations.

[0009] In one alternative implementation, to aid in quantifying the local curvature characteristics of the base surface, after reconstructing the discrete manifold triangular mesh, an operation is performed to extract the differential geometric properties of the mesh vertices, including: calculating the area of ​​the local control region of the mesh vertices using the hybrid Veronoi region method; calculating the average curvature normal vector of the mesh vertices using the discrete Laplace-Beltrami operator and based on the cotangent weight model; and calculating the Gaussian curvature of the mesh vertices based on the angle deficit principle.

[0010] In one optional implementation, the numerical setting of the global tolerance potential energy field follows a nonlinear penalty logic: when the absolute value of the algebraic symbol distance corresponding to a grid vertex is less than or equal to the process tolerance threshold, it is determined to be a qualified region, and the potential energy value of the grid vertex is set to zero; when the absolute value of the algebraic symbol distance corresponding to a grid vertex is greater than the process tolerance threshold, it is determined to be an out-of-tolerance region, and the potential energy value of the grid vertex is set to a positive value. This positive value increases nonlinearly as the difference between the algebraic symbol distance and the process tolerance threshold increases, thereby forming a numerical penalty for the marked path crossing the out-of-tolerance region, forcing the path to shift towards the region with qualified flatness.

[0011] In one optional implementation, to accommodate the allowable positioning deviation during on-site construction, the feasible region for anchor point drift is constructed as follows: a preset allowable positioning deviation radius is obtained, and spherical feasible regions for starting and ending points are constructed with the theoretical starting and ending points as the centers and the allowable positioning deviation radius as the radius, respectively. During path optimization, the starting and ending points of the optimal marked path are constrained to be located inside or on the boundary of the feasible regions for starting and ending points, respectively, rather than fixed to a single theoretical coordinate point.

[0012] In one optional implementation, the composite energy functional is defined as a weighted sum of the path tension energy term and the external tolerance potential energy term. The path tension energy term is obtained by calculating the sum of squared Euclidean distances between adjacent discrete nodes on the marked path, and is used to constrain the path length and straightness, reflecting the principles of shortest path and smoothness. The external tolerance potential energy term is obtained by accumulating the potential energy values ​​corresponding to all discrete nodes on the marked path in the global tolerance potential energy field, and is used to constrain the path's avoidance of surface defects. By adjusting the weighting coefficients, the emphasis of path optimization between geometric straightness and surface adaptability can be flexibly controlled.

[0013] In one optional implementation, the optimal marked path is solved using a projective gradient descent algorithm. The specific process includes: using the straight line segment connecting the theoretical start and end points as the initial path; iteratively updating the spatial coordinates of discrete nodes on the path, calculating the gradient vector of the composite energy functional with respect to the path node coordinates in each iteration, and updating the node coordinates along the negative gradient direction to reduce system energy; after updating the node coordinates, determining whether the start and end endpoints are within the feasible region of the anchor point drift; if they are outside the range, projecting the corresponding endpoint coordinates onto the boundary surface of the feasible region of the anchor point drift until the composite energy functional converges to its minimum value.

[0014] In one optional implementation, the specific logic for dividing the regions is determined based on Boolean interference relations and the directional attribute of the algebraic symbol distance: mesh vertices located inside the semantic envelope are divided into the effective regions, representing qualified base surfaces that do not require processing; mesh vertices located outside the semantic envelope and whose algebraic symbol distance indicates that they are located on the positive side (i.e., the convex side) of the ideal reference plane are divided into the polishing regions; and mesh vertices located outside the semantic envelope and whose algebraic symbol distance indicates that they are located on the negative side (i.e., the concave side) of the ideal reference plane are divided into the repair regions.

[0015] In one optional implementation, the digital marking model employs a structured data storage method. Specifically, this includes: discretizing and sampling the optimal marking path along its extension direction to construct an ordered sequence of multiple marker control points; and establishing a data structure for each marker control point in the ordered sequence, containing three-dimensional spatial absolute coordinates, a construction attribute status code, and quantitative operation parameters. The three-dimensional spatial absolute coordinates are the spatial coordinates of the marker control point at the current sampling position; the construction attribute status code stores the category identifier value corresponding to the marker control point based on the result of the region division; and the quantitative operation parameters store the construction quantity value. If the category indicated by the construction attribute status code is the grinding area or the repair area, the construction processing depth is written into the quantitative operation parameters; if the indicated category is the effective area, a zero value is written into the quantitative operation parameters.

[0016] This invention provides a digital modeling method for marking architectural decoration projects. It has the following beneficial effects: 1. This invention constructs a global tolerance potential energy field and an anchor point drift feasible region, and solves for the global extremum of the composite energy functional under this constraint. This invention can automatically plan the optimal marking path that balances geometric tension and substrate fit. Utilizing the gradient repulsion effect of the potential energy field, the path is guided to automatically avoid severely raised or recessed areas on the substrate. Simultaneously, the tension energy term maintains the geometric straightness of the path at the macroscopic level. Thus, at the mathematical level, a balance is achieved between the visual aesthetic requirements of decoration and the tolerance conditions of on-site construction. This effectively solves the technical problem of traditional rigid straight-line layout being unable to adapt to local flatness defects on the substrate, leading to rework.

[0017] 2. This invention generates a semantic envelope based on the optimal marking path and achieves precise digital quantification of hidden engineering quantities by calculating the Boolean interference relationship between the semantic envelope and the discrete manifold triangular mesh. Unlike the traditional discrete sampling method that relies on manual inspection rulers, this invention can perform automated mesh-level analysis of the entire base surface, accurately identify and classify effective areas, grinding areas, and repair areas, and directly output precise construction treatment depths. This provides on-site workers with fixed-point and quantitative digital construction instructions, significantly improving construction precision and effectively controlling material costs.

[0018] 3. This invention calculates the spatial positioning deviation vector between the optimal actual coordinates and the theoretical design coordinates, and feeds this deviation data back to the Building Information Model (BIM) to correct component properties. This invention establishes a closed-loop data flow from digital design to digital construction and back to digital archiving. This dynamic correction mechanism based on measured data ensures that the geometric topology in the BIM remains consistent with the physical site after tolerance optimization, eliminating geometric conflicts between theoretical design data and the actual environment. This provides a reliable positioning benchmark for the precise installation of subsequent prefabricated components such as keels and decorative panels, avoiding accumulated installation errors. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating a digital modeling method for marking architectural decoration projects according to an embodiment of the present invention. Figure 2 This is a hardware operating environment and system architecture diagram of one embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the optimization principle of the anchor point drift feasible region and the optimal marker path according to an embodiment of the present invention; Figure 4 This is a logical cross-sectional diagram of the semantic envelope construction and base plane region partitioning according to an embodiment of the present invention. Detailed Implementation

[0020] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] See attached document Figure 1 This invention provides a digital modeling method for marking architectural decoration projects. This method is executed by a computer processor and mainly includes the following steps: S100: Obtain the original point cloud data of the building base surface, reconstruct the original point cloud data into a discrete manifold triangular mesh, and calculate the geometric feature parameters of each vertex on the discrete manifold triangular mesh. S200 determines the tolerance threshold according to the preset decoration process standards, and constructs a global tolerance potential energy field based on the geometric deviation of the discrete manifold triangular mesh relative to the ideal reference surface. S300: Based on the theoretical start and end points determined by the design drawings, define the feasible region of anchor point drift, construct a composite energy functional including tension energy term and tolerance potential energy term, solve for the minimum value of composite energy functional under the constraint of the feasible region of anchor point drift, and obtain the optimal marking path and the optimal start and end point coordinates. S400 generates a semantic envelope with the optimal marking path as the central axis, calculates the Boolean interference relationship between the discrete manifold triangular mesh and the semantic envelope, and divides the region on the discrete manifold triangular mesh into effective regions and conflict regions. S500 generates a digital marker model containing construction attributes based on the regional division results, and feeds back the optimal start and end point coordinates to the building information model to update the positioning data.

[0022] In step S100, the original point cloud of the building base surface is reconstructed into a discrete manifold triangular mesh, and the differential geometric properties of the vertices of the discrete manifold triangular mesh are extracted. In this step, the acquisition and preprocessing of point cloud data are performed first.

[0023] See attached document Figure 2 The processor acquires the base space data of the building decoration site using a high-precision 3D laser scanning device, forming a raw point cloud dataset. The raw point cloud dataset is denoted as... The original point cloud dataset contains A set of discrete three-dimensional coordinate points. Each original coordinate point in the set. It includes the x-coordinate, y-coordinate, and vertical coordinate values ​​in a spatial rectangular coordinate system.

[0024] Due to dust interference and equipment measurement errors in the on-site environment, the original point cloud dataset inevitably contains outliers and noise. The processor employs a statistical filtering algorithm to clean the original point cloud dataset. For any target point in the original point cloud dataset... The system searches for the target point that is closest to it in three-dimensional space. Find the nearest neighbor points and calculate the distance between the target point and the target point. The average Euclidean distance between the nearest neighbors is denoted as . .

[0025] The processor iterates through all coordinate points to calculate the global mean of the average Euclidean distance. and global standard deviation A distance threshold is set based on the normal distribution assumption. Distance threshold The calculation formula is as follows: ; in, This represents the standard deviation factor, used to adjust the sensitivity of the filter. The processor calculates the average Euclidean distance for each target point. With distance threshold Perform a comparison. When Greater than the distance threshold When the target point is identified as an outlier and is removed from the dataset; when Less than or equal to the distance threshold At that time, retain the target point.

[0026] Point cloud data density after noise cleaning is often too high, and direct processing would consume excessive computational resources. The processor employs a voxel grid downsampling method to sparsify the data. The system constructs a 3D model with sides of length... The voxel grid network is used. For all point cloud data points falling within the same voxel grid, the processor calculates the geometric centroid of these points as the unique representative point of that voxel grid. The formula for calculating the coordinates of the geometric centroid is as follows: ; in, This represents the coordinates of the retained points after downsampling. This represents the number of original points contained within the current voxel grid. Represents the current voxel grid. The coordinate vectors of the original points.

[0027] After the above denoising and downsampling processes, the system outputs a preprocessed clean point cloud dataset. Clean point cloud dataset This will serve as the base data source for the subsequent construction of the triangular mesh.

[0028] Based on the preprocessed clean point cloud dataset The processor then performs a reconstruction operation of the discrete manifold triangular mesh. This process aims to establish topological connections between points, transforming the discrete set of spatial points into a geometric model with continuous surface features. The processor first computes the clean point cloud dataset. The normal vector information of each point in the dataset is used to determine the local orientation of the surface. For any point in the dataset... Select its spatial neighborhood Construct a local covariance matrix from its nearest neighbors. Covariance matrix The calculation formula is as follows: ; in, The geometric center vector representing the coordinates of all points in the neighborhood. This represents the matrix transpose operation.

[0029] Processor on covariance matrix Eigenvalue decomposition yields three non-negative eigenvalues ​​and their corresponding eigenvectors. The eigenvector corresponding to the smallest eigenvalue is determined as a point. normal vector To ensure global consistency of the normal vector direction, the system constructs a Riemann graph and uses the minimum spanning tree algorithm to propagate the normal vector direction to the entire domain, ensuring that all normal vectors point uniformly to the outside of the base plane.

[0030] After acquiring point cloud data with normal vectors, the processor uses a Poisson surface reconstruction algorithm to construct implicit surface functions. The system defines a vector field in space. , making the vector field The numerical values ​​at the sampling points and the calculated normal vector A match is found. The system solves the following Poisson equation to obtain the indicator function. : ; in, Represents the Laplace operator, Represents the divergence operator. Represents any coordinate position in three-dimensional space. Indicator function. Used to characterize the probability distribution of spatial points located inside or outside the model. The processor extracts indicator functions. The isosurfaces are used to generate the triangular mesh surface. An isosurface threshold is set, and the moving cube algorithm is used to divide the spatial voxels into internal and external regions. The intersections of the isosurfaces and voxel edges are calculated using linear interpolation. Connecting these intersections forms triangular patches, thus generating the initial triangular mesh model.

[0031] To ensure the generated mesh satisfies manifold properties, the processor performs topology checks and repairs on the initial triangular mesh model. The system detects and removes all non-manifold edges and vertices. A non-manifold edge is defined as an edge shared by more than two triangles, and a non-manifold vertex is defined as a vertex whose neighboring triangles cannot form a single connected sector.

[0032] After the above reconstruction and repair steps, the system outputs a discrete manifold triangular mesh. Discrete manifold triangular mesh From the set of vertices With triangular facet set It consists of a topological disk in which any edge is shared by at most two triangular facets, and the triangular facets surrounding any vertex form a closed or open topological disk, providing a continuity basis for subsequent differential geometric property calculations.

[0033] After obtaining the discrete manifold triangular mesh, the processor further performs the extraction of differential geometric properties of the mesh vertices to quantify the local curvature and flatness characteristics of the building base surface. The calculation of differential geometric properties mainly focuses on two core parameters: mean curvature and Gaussian curvature. These two parameters together describe the geometry of the base surface at the microscale.

[0034] To perform accurate numerical differentiation calculations in discrete space, the system first considers each vertex in the discrete manifold triangular mesh. Define the area of ​​the local control region Area of ​​the local control area The calculation is performed using the hybrid Voronoi region method. For calculations around the vertices... For any adjacent triangle, if the triangle is non-obtuse, calculate the area of ​​the Voronoi region enclosed by the circumcenter of the triangle to the midpoints of each side as the contribution value; if the adjacent triangle contains obtuse angles, use half the area of ​​the centroid region of the triangle as the contribution value. [The text then abruptly shifts to a different topic:] ...vertices... The contribution values ​​of all adjacent triangles are summed to obtain the area of ​​the local control region used for normalization calculation. .

[0035] The processor uses the discrete Laplace-Beltrami operator to compute vertices. Mean curvature normal vector The calculation process is based on a cotangent weight model, designed to capture the average curvature of the mesh surface along the normal direction. The average curvature normal vector... The calculation formula is as follows: ; in, Representing the vertex The set of vertices in the ring neighborhood, This represents the coordinate vector of adjacent vertices within the neighborhood. and Representing the connected vertices With vertex The angles of the interior angles opposite the common side of two adjacent triangles. Represents the cotangent trigonometric function. The final scalar value of the mean curvature. Take the mean curvature normal vector Half the length of the mold.

[0036] The processor calculates vertices based on the principle of angle loss. Gaussian curvature Gaussian curvature characterizes the intrinsic bending property of a surface at a point and is used to distinguish convex, concave, and saddle-point regions on a base surface. The calculation formula is as follows: ; in, Represents pi (π) Representatives and Vertices The total number of directly connected adjacent triangles. Representing the Adjacent triangles at vertex The interior angle value at that location. The system will calculate the average curvature. With Gaussian curvature The differential geometric eigenvectors that form the vertices are combined. These eigenvectors reveal the second-order geometric rate of change of the discrete manifold triangular mesh surface, providing a precise geometric quantitative basis for subsequent steps to determine whether the base surface meets the flatness requirements and to construct the potential energy field.

[0037] In step S200, firstly, based on the construction and acceptance specifications for building decoration projects, the maximum allowable surface flatness deviation for the current construction process is determined. This maximum surface flatness deviation is defined as the tolerance threshold, denoted as... Tolerance threshold It is the quantitative boundary for determining whether the building base meets the requirements of subsequent processes, and directly determines the range of the zero potential energy interval of the potential energy field in subsequent calculations.

[0038] The processor constructs an ideal reference plane based on the spatial positioning data in the design drawings. This represents the spatial geometric position of the building surface under an ideal, error-free design. Ideal reference plane. The mathematical definition is based on the point-normal form equation, which is expressed as follows: ; in, The unit normal vector representing the ideal reference plane is used to determine the spatial orientation of the plane; The constant term in the plane equation represents the distance offset of the plane relative to the origin. It represents the spatial position vector of any point on the plane.

[0039] Based on the constructed ideal reference plane The processor traverses the discrete manifold triangular mesh. Each vertex in Calculate vertices Relative to the ideal reference plane algebraic symbolic distance Algebraic symbolic distance This is used to accurately quantify the degree and direction of deviation of discrete points on the base surface from their designed positions. The calculation formula is as follows: ; in, Represents the first triangular mesh of a discrete manifold The coordinate vectors of the vertices. The calculated algebraic signed distance. It has a clear geometric and physical meaning: if A positive value indicates a vertex. Located in the ideal reference plane The normal positive side corresponds to the raised area on the building base; if A negative value indicates a vertex Located in the ideal reference plane The negative normal side corresponds to the recessed area on the building base. Algebraic symbolic distance. The absolute value of is the geometric deviation at that point.

[0040] After obtaining the algebraic sign distance of each vertex on the discrete manifold triangular mesh, the processor further constructs a global tolerance potential energy field defined on the surface of the discrete manifold triangular mesh. The global tolerance potential energy field assigns a specific potential energy value to each location on the mesh, and the magnitude of the potential energy value represents the correction cost caused by the unevenness of the base surface when construction marking is carried out at the corresponding location.

[0041] The construction process incorporates the algebraic symbolic distance obtained from previous steps and a preset tolerance threshold. For any vertex on the mesh, if the absolute value of the algebraic symbolic distance corresponding to the vertex is less than or equal to the tolerance threshold, the vertex is determined to be in a qualified region. The geometry within the qualified region meets the construction process requirements, and the corresponding potential energy value is set to zero.

[0042] A vertex is considered to be in an out-of-tolerance region when the absolute value of the algebraic symbol distance to the vertex exceeds the tolerance threshold. An out-of-tolerance region means that the degree of bulging or concavity of the basal surface exceeds the allowable range of the manufacturing process. The processor assigns a positive potential energy to vertices in the out-of-tolerance region, and the potential energy value increases non-linearly as the difference between the algebraic symbol distance and the tolerance threshold increases.

[0043] The processor uses a piecewise function approach to compute the first... Potential energy values ​​at each vertex The calculation formula is as follows: ; in, Indicates the first The absolute value of the algebraic sign distance of each vertex; Indicates the tolerance threshold; This represents the potential energy penalty coefficient.

[0044] Potential energy penalty coefficient A constant greater than zero is set to adjust the sensitivity of the global tolerance potential energy field to out-of-tolerance geometric features. A higher potential energy penalty coefficient causes the potential energy value in the out-of-tolerance region to rise sharply with the increase of deviation. In subsequent path optimization calculations, the high potential energy region creates a numerical repulsion effect, causing the calculated marked paths to preferentially distribute in flat areas within the tolerance threshold range, thus mathematically suppressing the amount of construction grinding and repair.

[0045] See attached document Figure 3 In step S300, the processor first parses and obtains the theoretical starting point coordinates and theoretical ending point coordinates corresponding to the current marking task based on the architectural decoration design drawings or building information model data. The theoretical starting point coordinates are denoted as... The theoretical termination point coordinates are marked as These two coordinate points represent the predetermined positions of the decoration marking lines within the architectural space under ideal design conditions.

[0046] The processor reads the construction design specification parameters associated with the current marking task, determines the allowable positioning deviation radius, and denotes it as... The allowable positioning deviation radius defines the maximum spatial displacement that the actual construction marker point can undergo relative to the theoretical design point. This parameter is set based on the accuracy requirements of specific construction nodes such as the positioning of non-load-bearing walls and the starting point of floor mosaic patterns.

[0047] Based on the acquired theoretical coordinates and the allowable positioning deviation radius, the processor constructs a feasible region for anchor point drift in three-dimensional space. The feasible region for anchor point drift includes the feasible region for starting point drift. Feasible region for drift with endpoint These two regions define the spatial range within which the starting and ending points of the marker lines can be moved during subsequent optimization calculations.

[0048] Feasible region for starting point drift Defined as starting from the theoretical point of view Centered on the sphere, with the allowable positioning deviation radius A set of points on a spatial sphere with radius . Feasible region for endpoint drift. Defined as the theoretical termination point Centered on the sphere, with the allowable positioning deviation radius Let be the set of points on a spatial sphere with radius . The calculation formula is as follows: ; ; Among the above, Representing three-dimensional real space Any coordinate point vector in the vector, This represents the Euclidean norm operation. The constructed feasible region for anchor point drift provides a constrained variable space for subsequent path optimization algorithms, enabling the final generated marker line to adapt to complex base surface geometry features through minute displacements of the endpoints, while meeting design specifications.

[0049] Building upon this, the processor constructs a composite energy functional comprising tension energy and tolerance potential energy terms, mathematically and quantitatively describing the physical equilibrium state of the marked path under geometric tension and datum tolerance constraints. Composite Energy Function Defined as a marker path tension energy With tolerance potential energy The weighted sum.

[0050] In order to perform numerical calculations, the processor will select the marked path to be solved. Discretization into inclusion An ordered sequence of spatial nodes. The ordered sequence is denoted as . ,in Corresponding to the starting point, Corresponding termination point, intermediate node Distributed along a three-dimensional spatial path between the starting point and the ending point.

[0051] Tension energy term This simulation demonstrates the geometric properties of a physical wire tending towards a straight line or geodesic when taut at both ends. The processor uses either the Discrete Hooke's Law model or the Dirichlet energy model to calculate the tension energy, which is the sum of the squares of the Euclidean distances between adjacent nodes on the path. A smaller tension energy value indicates a shorter and straighter path.

[0052] Tolerance potential term Introducing the global tolerance potential energy field data constructed in the previous steps. The potential energy value corresponding to each discrete node on the processor's computation path within the global tolerance potential energy field. The values ​​are then summed. This value reflects the cumulative cost of the entire marked path traversing the out-of-tolerance region of the datum. For spatial nodes not directly located on the mesh surface, the processor projects them onto the surface of the discrete manifold triangular mesh to obtain the potential energy value at the projection point.

[0053] Composite energy functional The complete calculation formula is as follows: ; in, This represents the tension weighting coefficient. This represents the potential energy weighting coefficient. This represents the operation of vector magnitude. Indicates the first The base surface potential energy values ​​corresponding to each path node. These two weighting coefficients are used to adjust the model's emphasis on line straightness and construction avoidance requirements. In practical applications, by increasing... Compared to The ratio of [value] can drive the generated path to bend more aggressively to avoid high-potential-energy protrusions on the wall.

[0054] After constructing the composite energy functional, the processor performs a constraint-based global extremum solution. The solution process aims to find an optimal sequence of spatial node coordinates that minimizes the composite energy functional across the entire domain, while ensuring that the starting and ending nodes of the path always lie within the feasible region of anchor point drift defined in the previous steps.

[0055] The processor employs either the projective gradient descent algorithm or the variational iterative algorithm for numerical solution. First, the initial state of the marked path is set as a straight line segment connecting the theoretical starting point and the theoretical ending point. Then, the processor calculates the gradient vector of the composite energy functional with respect to the coordinates of each path node. The gradient vector indicates the direction of the fastest energy decrease and includes the contractile force component from the tension energy term and the repulsive force component from the tolerance potential energy term.

[0056] In each step of the iterative computation, the processor updates the spatial coordinates of the path nodes according to the gradient direction. For nodes in the middle of the path, the coordinate updates follow the negative gradient direction, gradually moving them away from the high-potential region on the base surface while maintaining geometric smoothness. For the starting and ending nodes, after performing the gradient update, the processor needs to verify whether the node coordinates exceed the boundary of the feasible region for anchor point drift.

[0057] If the updated start or end node coordinates are located outside the drift feasible region, the processor performs a spatial projection operation to forcibly map the node coordinates back onto the boundary surface of the drift feasible region, ensuring that the boundary constraints are always satisfied. Optimal Marking Path The mathematical model for solving this problem is expressed as follows: ; ; in, This represents the set of variable values ​​that minimize the objective function. Indicates being bound by; and These represent the starting and ending nodes of the path, respectively. and These represent the feasible regions for drifting from the starting point and the feasible regions for drifting from the ending point, respectively. This represents any candidate marked path under the anchor point drift feasible region constraint.

[0058] As the number of iterations increases, the composite energy functional gradually converges. When the energy difference between two consecutive iterations is less than a preset convergence threshold, the processor terminates the iteration process. The node sequence obtained at this point is the optimal marking path. The optimal marking path appears in three-dimensional space as a taut curve with slight local curvature. This curve automatically avoids severe basal surface bulges or depressions, and its endpoints are within the design-allowed error range.

[0059] See attached document Figure 4 In step S400, the processor generates a semantic envelope for spatial interferometry analysis based on the optimal marking path calculated in the previous steps. The semantic envelope is a three-dimensional tubular geometric region extending along the optimal marking path; its spatial shape intuitively reflects the allowable range of base surface flatness for the construction process. The processor uses the optimal marking path... Using the central axis as the tolerance threshold determined in step S200 Using the radius parameter, construct a three-dimensional point set. The semantic envelope is denoted as... Its mathematical definition is the set of all spatial points whose shortest Euclidean distance to the optimal marked path is less than or equal to the tolerance threshold.

[0060] Semantic envelope The set expression is as follows: ; in, Represents any coordinate point in three-dimensional space; This indicates the points on the optimal marked path in the parameters. The following position coordinates; Representing a spatial point To path curve The shortest distance.

[0061] In discrete numerical computation, the optimal marked path consists of a series of straight line segments connecting adjacent nodes. The processor constructs the semantic envelope as a Boolean union of cylinders with each straight line segment as its axis and a tolerance threshold as its radius, and spheres with each path node as its center and a tolerance threshold as its radius. The resulting closed geometry forms a virtual pipeline in three-dimensional space with the optimal path as its core. The spatial region inside this pipeline represents the acceptable spatial range that meets the flatness acceptance requirements.

[0062] The generated semantic envelope establishes the acceptable spatial range, and the processor further calculates the Boolean interference relationship between the discrete manifold triangular mesh and the semantic envelope. By traversing each vertex in the discrete manifold triangular mesh, the internal and external positional relationship of the mesh vertex relative to the semantic envelope is determined, thereby dividing the building base surface into construction areas with different attributes.

[0063] The set of valid regions is defined as the set of mesh vertices located inside the semantic envelope. Let the set of valid regions be denoted as... For the mesh vertices within the set, the geometric position deviation is within the tolerance threshold and no physical processing is required.

[0064] The set of polishing regions is defined as the set of mesh vertices located outside the semantic envelope, and whose algebraic symbolic distance indicates they lie on the positive side of the normal to the ideal reference plane. Let the set of polishing regions be... The grinding area corresponds to the protruding parts on the building base surface, and the degree of protrusion exceeds the tolerance threshold, so the excess material must be removed by physical grinding.

[0065] The set of patchable regions is defined as the set of mesh vertices located outside the semantic envelope, and whose algebraic symbolic distance indicates they lie on the negative side of the normal to the ideal reference plane. Let the set of patchable regions be denoted as... The repair area set corresponds to deep depressions on the building base surface, which must be filled and repaired using grouting or plastering methods. The mathematical logic for area classification is as follows: ; ; in, Represents grid vertices; Represents the semantic envelope; Represents the algebraic symbolic distance of a mesh vertex relative to an ideal reference plane; This represents the logical AND operation; This represents the global set of vertices on the reconstructed building base surface, i.e., the total set of all mesh vertices.

[0066] For each vertex to be processed in the set of grinding and repair areas, the processor calculates the specific processing depth. The processing depth is denoted as... The construction treatment depth is equal to the absolute value of the algebraic symbolic distance between the vertices to be treated minus the tolerance threshold. The calculation formula is as follows: ; in, Represents the absolute value of the distance between algebraic symbols; This represents the tolerance threshold. The calculated construction treatment depth value accurately quantifies the thickness of concrete that needs to be removed or the thickness of mortar that needs to be backfilled by on-site construction personnel, realizing digital quantitative analysis of the amount of concealed works.

[0067] In step S500, the processor first performs the construction and output operation of the composite marker data structure. Based on the region division results and construction processing depth data obtained in the previous steps, the processor transforms the optimal marker path of pure geometry into a digital marker model containing rich construction semantics. The digital marker model is organized into an ordered sequence of discrete marker control points, and each unit in the ordered sequence encapsulates spatial positioning information and process processing instructions.

[0068] The processor defines the data structure for marking control points as a triplet. Let the first... The data tuple for each marked control point is: Data tuples The mathematical expression is as follows: ; in, It represents the absolute coordinate vector of the marked control point in three-dimensional space, used to guide construction layout equipment or augmented reality display devices to make accurate projections; This indicates the construction attribute status code, used to identify the current location of the base surface quality status; This represents quantitative operational parameters, used to indicate specific construction and processing quantities.

[0069] Construction attribute status codes The definition is based on enumerated values. When the discrete manifold triangular mesh region corresponding to the marked control point belongs to the valid region set... hour, Assign a value to the standard state value (e.g., 0), and then quantize the operation parameters. Setting it to zero indicates that the flatness of the base surface at this location is acceptable, and only routine chalk line marking is required.

[0070] When the grid area corresponding to the marked control point belongs to the grinding area set hour, Assign a value to the grinding status (e.g., 1), and then quantify the operation parameters. The value is assigned to the construction treatment depth calculated in the previous steps. The value directly corresponds to the thickness of concrete that needs to be removed on site.

[0071] When the grid area corresponding to the marked control point belongs to the repair area set hour, Assign a value to the repair status (e.g., -1), and then quantize the operation parameters. Also related to the depth of construction treatment The numerical value indicates the thickness of the material that needs to be backfilled.

[0072] Based on this, the processor further performs reverse correction operations on the positioning data, establishing a closed-loop feedback mechanism between the physical state of the construction site and the digital state of the building information model. The processor calculates the optimal marking path from the previous steps. Extract the coordinates of the starting node and the ending node, and define them as the optimal actual starting point. and the optimal actual termination point These two coordinate points represent the precise location in physical space where the marker line should actually be positioned, taking into account both the geometry of the base surface and construction tolerance constraints. The processor calculates the spatial positioning deviation vector based on the optimal actual coordinates and the theoretical design coordinates called in step S300. The spatial positioning deviation vector is denoted as... The formula for calculating the positioning deviation vector for the starting point position is as follows: ; in, This represents the starting point coordinate vector after path optimization; This represents the coordinate vector of the theoretical starting point originally defined in the Building Information Model (BIM). Similarly, the positioning deviation vector of the ending point is calculated. These two vectors precisely quantify the extent of adjustments made to the construction plan relative to the original design.

[0073] The processor uses the Building Information Modeling (BIM) application programming interface (API) to identify the unique identifier of the component object associated with the current decoration marking task. The processor then generates parameter update instructions, replacing the original theoretical coordinate data with the optimal actual starting point. and the optimal actual termination point and the positioning deviation vector This information is written as attribute metadata into the component's historical change record.

[0074] Through the aforementioned data feedback operation, the geometric topology in the Building Information Model (BIM) is automatically updated to ensure consistency with the actual geometric state of the construction site. Subsequent finishing processes (such as keel installation and panel laying) can directly obtain the actual positioning benchmark after tolerance optimization when reading the BIM, avoiding the cumulative errors caused by using theoretical design data, and realizing the full lifecycle data flow from digital design to digital construction and back to digital archives.

[0075] To implement the aforementioned digital modeling method for marking architectural decoration projects, this embodiment also provides a computer device. The computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor.

[0076] When the processor executes the computer program therein, it implements the functions of S100 to S500 and their sub-steps.

[0077] Specifically, the computer equipment is physically connected to a 3D laser scanner (or LiDAR sensor) as a data acquisition terminal. The 3D laser scanner is used to acquire raw point cloud data of the building's base surface and transmits the raw point cloud data to the processor via a high-speed data bus.

[0078] The processor is configured as a high-performance computing unit, used to perform the reconstruction of discrete manifold triangular meshes, extraction of differential geometric properties, construction of global tolerance potential fields, and extreme value solving of composite energy functionals.

[0079] The computer equipment is also connected to a visual interactive terminal or an augmented reality (AR) projection device.

[0080] Visual interactive terminal: used to display the generated digital marking model, showing the distribution of effective areas, grinding areas and repair areas in the form of 3D rendering, and using heat map colors to indicate the depth of construction treatment.

[0081] Data Interface: Used to establish bidirectional communication with the Building Information Modeling (BIM) server, feeding back the spatial positioning data and positioning deviation data of the calculated optimal marking path to the BIM database to update the attribute parameters of component objects.

[0082] Furthermore, embodiments of the present invention also provide a computer-readable storage medium, wherein a computer program is stored on the storage medium, and the computer program, when executed by a processor, implements the steps of the above-described digital modeling method for marking architectural decoration projects.

Claims

1. A digital modeling method for marking architectural decoration projects, characterized in that, Includes the following steps: Acquire the original point cloud data of the building base surface and reconstruct it into a discrete manifold triangular mesh; The discrete manifold triangular mesh is fitted with a plane to determine the ideal reference plane. Based on the preset process tolerance threshold, the algebraic sign distance between the mesh vertices and the ideal reference plane is calculated to construct a global tolerance potential energy field. The theoretical start and end points are obtained from the building information model. Anchor point drift feasible region is constructed based on the theoretical start and end points. Composite energy functional is established in combination with the global tolerance potential energy field. The optimal marking path that minimizes the composite energy functional is solved under the constraint of the anchor point drift feasible region. The optimal marking path is used as the central axis to generate a semantic envelope. The Boolean interference relationship between the semantic envelope and the discrete manifold triangular mesh is calculated. Based on the Boolean interference relationship and the directional attribute of the algebraic symbol distance, the region is divided and the construction processing depth is calculated. Based on the results of the area division and the construction processing depth, a digital marking model containing construction attributes is generated to guide on-site operations. Simultaneously, the spatial coordinates of the optimal marking path are used to correct the component positioning data corresponding to the building decoration engineering markings in the building information model.

2. The digital modeling method for marking architectural decoration projects according to claim 1, characterized in that, The step of acquiring the original point cloud data of the building base surface and reconstructing it into a discrete manifold triangular mesh includes: The acquired raw point cloud data is subjected to statistical denoising processing, the average Euclidean distance between the target point and its nearest neighbors is calculated, and outlier noise points exceeding the distance threshold are removed. The denoised point cloud data is downsampled using a voxel grid. For point cloud data falling within the same voxel grid, the geometric centroid is calculated as the representative point of the voxel grid. Calculate the normal vector of the downsampled point cloud data, and use the minimum spanning tree algorithm to unify the direction of the normal vector to obtain a clean point cloud dataset with normal vectors.

3. The digital modeling method for marking architectural decoration projects according to claim 2, characterized in that, Also includes: An implicit surface function is constructed based on the clean point cloud dataset with normal vectors using the Poisson surface reconstruction algorithm. The indicator function is obtained by solving the Poisson equation, and the isosurface of the indicator function is extracted to generate an initial triangular mesh. A topology check is performed on the initial triangular mesh to detect and remove non-manifold edges shared by more than two triangles and non-manifold vertices that cannot form a single connected sector, thus obtaining the discrete manifold triangular mesh.

4. The digital modeling method for marking architectural decoration projects according to claim 1, characterized in that, After reconstructing the mesh into a discrete manifold triangular mesh, the process also includes extracting the differential geometric properties of the mesh vertices: The area of ​​the local control region at the mesh vertices is calculated using the hybrid Veronoy region method; The average curvature normal vector of the mesh vertices is calculated using the discrete Laplace-Beltramian operator and based on the cotangent weight model. Calculate the Gaussian curvature of the grid vertices based on the angle loss principle; The average curvature normal vector and the Gaussian curvature are used to assist in quantifying the local bending characteristics of the base surface.

5. The digital modeling method for marking architectural decoration projects according to claim 1, characterized in that, In the step of constructing the global tolerance potential energy field, the numerical setting logic of the global tolerance potential energy field is as follows: When the absolute value of the distance between the algebraic symbols corresponding to a grid vertex is less than or equal to the process tolerance threshold, the potential energy value of the vertex is set to zero, corresponding to a qualified area. When the absolute value of the distance between the algebraic symbols corresponding to the grid vertex is greater than the process tolerance threshold, the potential energy value of the vertex is set to a positive value, and the positive value increases non-linearly as the difference between the distance between the algebraic symbols and the process tolerance threshold increases, forming a numerical penalty for the marked path to pass through the out-of-tolerance region.

6. The digital modeling method for marking architectural decoration projects according to claim 1, characterized in that, The step of constructing the feasible region for anchor point drift based on the theoretical start and end points includes: Obtain the preset allowable positioning deviation radius; Construct a feasible region for starting point drift by taking the theoretical starting point as the center of the sphere and the allowable positioning deviation radius as the radius. Construct a feasible region for the drift of the termination point by taking the theoretical termination point as the center of the sphere and the allowable positioning deviation radius as the radius. The starting and ending endpoints of the optimal marking path are constrained to be located inside or on the boundary of the feasible region of the starting point drift and the feasible region of the ending point drift, respectively.

7. The digital modeling method for marking architectural decoration projects according to claim 1, characterized in that, The steps for establishing a composite energy functional based on the global tolerance potential field include: The composite energy functional is defined as a weighted sum of the path tension energy term and the external tolerance potential term; The path tension energy term is obtained by calculating the sum of squared Euclidean distances between adjacent discrete nodes on the marked path, and is used to constrain the length and straightness of the path. The external tolerance potential energy term is obtained by accumulating the potential energy values ​​of all discrete nodes on the marked path in the global tolerance potential energy field, and is used to constrain the path to avoid base surface defects. The focus of path optimization is controlled by adjusting the weighting coefficients.

8. The digital modeling method for marking architectural decoration projects according to claim 7, characterized in that, The steps of finding the optimal marker path that minimizes the composite energy functional under the constraints of the feasible region of anchor point drift include: Use the straight line segment connecting the theoretical start and end points as the initial path; The spatial coordinates of discrete nodes on the path are iteratively updated using the projected gradient descent algorithm. In each iteration, the gradient vector of the composite energy functional with respect to the coordinates of the path nodes is calculated, and the node coordinates are updated along the negative gradient direction. After updating the node coordinates, it is determined whether the starting endpoint and the ending endpoint are located within the feasible region of the anchor point drift. If they are outside the range, the corresponding endpoint coordinates are projected and mapped to the boundary surface of the feasible region of the anchor point drift until the composite energy functional converges.

9. The digital modeling method for marking architectural decoration projects according to claim 1, characterized in that, The step of dividing the region based on the Boolean interference relation and the directional attribute of the algebraic symbol distance includes: The grid vertices located inside the semantic envelope are divided into the effective regions; The mesh vertices located outside the semantic envelope and whose algebraic symbol distance indicators are located on the positive side of the normal to the ideal reference plane are divided into the polishing regions; The mesh vertices located outside the semantic envelope and whose algebraic symbol distance indicators are located on the negative side of the ideal reference plane are designated as the patching region.

10. The digital modeling method for marking architectural decoration projects according to claim 1, characterized in that, The step of generating a digital marker model containing construction attributes based on the results of the region division and the construction processing depth includes: The optimal marking path is discretized and sampled along its path extension direction to construct an ordered sequence composed of multiple marking control points; For each marked control point in the ordered sequence, establish a data structure containing three-dimensional spatial absolute coordinates, construction attribute status codes, and quantitative operation parameters; Wherein, the three-dimensional spatial absolute coordinates are the spatial coordinates of the marked control point at the current sampling position; the construction attribute status code is used to store the category identifier value corresponding to the marked control point according to the result of the division of the region; The quantitative operation parameters are used to store the construction quantity values. If the category indicated by the construction attribute status code is the grinding area or the repair area, the construction processing depth is written into the quantitative operation parameters. If the indicated category is the valid area, a zero value is written into the quantitative operation parameters.

Images