A porous structure design method based on implicit B-spline

The discrete distance field of the porous structure is fitted through the implicit B-spline function to generate a continuous distance distribution field, which solves the problems of mechanical and biological performance balance and storage efficiency in complex porous structure design, and realizes efficient porous structure design and manufacturing.

CN114970160BActive Publication Date: 2024-11-19ZHEJIANG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210592680.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-05-27
Publication Date
2024-11-19
Estimated Expiration
2042-05-27

AI Technical Summary

Technical Problem

The prior art is difficult to effectively design and regulate complex shape porous structures, which makes it difficult to balance mechanical properties and biological properties. At the same time, traditional STL file representation methods lead to high memory consumption.

Method used

Using a porous structure design method based on implicit B splines, the discrete distance field file of the porous structure is input, and iteratively fitting is used using the three-variable implicit B spline function to generate a continuous distance distribution field, and stored in ITDF file format. Realize editing and manufacturing of porous structures.

Benefits of technology

It reduces the storage space requirement, improves the controllability and expression ability of the porous structure, can effectively regulate the porosity and shape of the structure, and maintains good mechanical and biological performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114970160B_ABST
    Figure CN114970160B_ABST
Patent Text Reader

Abstract

The present invention discloses a porous structure design method based on implicit B-splines. A porous structure discrete distance field file of a real porous structure is input, including the number of grids along three directions in the discrete distance field and the position distance field value of each grid; a three-variable implicit B-spline function is selected to iteratively fit the position distance field value of each grid in the discrete distance field to obtain a continuous distance distribution field; according to the structural type of the porous structure input by the user, the porous structure in space is obtained in combination with the continuous distance distribution field processing; the porous structure is stored in the ITDF file format. The method of the present invention ensures the internal smoothness and controllability of the porous structure, can effectively regulate the porosity and shape of the porous structure, and can greatly reduce the storage space.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a computer-aided design method, in particular to a porous structure design method based on implicit B-splines. Background Art

[0002] Porous structures are widely present in various tissues, such as biomembranes, butterfly scales, and bone tissue. Porous structures are widely used in various fields due to their high internal connectivity, light weight, and high specific surface area. In the aerospace field, honeycomb structures are often used in wing design due to the high strength and low mass of porous structures; in the chemical industry, porous structures with high specific surface area can increase the contact area between catalysts and chemical materials and are used as catalyst carriers; in bone tissue engineering, the high specific surface area of ​​porous structures provides a wide space for cell attachment and growth, and the internally connected pores facilitate the exchange of nutrients and the metabolism of waste, and are widely used in bone scaffold design.

[0003] Due to the complex pore shapes, complex structural distribution and small pore size inside the porous structure, how to effectively design and manufacture the porous structure is a hot topic in current research, and many ways of representing porous structures have been born. Traditional CAD porous structure units such as BCC (body-centered cubic) units (see Maskery I, Hussey A, Panesar A, et al. An investigation into reinforced and functionally graded lattice structures [J]. Journal of Cellular Plastics, 2017, 53 (2): 151-165.) and FCC (face-centered cubic) units (see Ahmadi SM, Campoli G, Yavari SA, et al. Mechanical behavior of regular open-cell porous biomaterials made of diamond lattice unit cells [J]. Journal of the mechanical behavior of biomedical Materials, 2014, 34: 106-115.), its main feature is that it is represented by simple geometric shapes. Due to its excellent mechanical properties, it has been widely used in various engineering fields; Voronoi porous unit is a porous structure unit obtained by discretizing the three-dimensional space into Voronoi units, treating the vertices as the connection nodes of the porous structure and the edges as the rods of the porous structure. It has good mechanical and biological properties and is widely used in bone tissue engineering (see Martínez J, Hornus S, Song H, et al. Polyhedral Voronoi diagrams for additive manufacturing [J]. ACM Transactions on Graphics (TOG), 2018, 37 (4): 1-15.); Recently, the three-dimensional periodic minimal surface (TPMS) unit is a current research hotspot. TPMS is a periodic surface represented by an implicit function, which can be infinitely extended to the entire space in three directions. Many studies have shown that TPMS units are widely present in biological structures such as sea urchin biofilms and butterfly scales. Due to its implicit representation characteristics, the TPMS unit is highly controllable, and due to its bionic characteristics, it has good biological and mechanical properties, so it is widely used in various engineering fields.(See Melchels FPW, Bertoldi K, Gabbrielli R, et al. Mathematically defined tissue engineering scaffold architectures prepared by stereolithography [J]. Biomaterials, 2010, 31(27):6909-6916.).

[0004] Since biological tissue is an extremely complex structure, using simple porous units to represent complex porous structures has the problem of insufficient expression. Since there is a trade-off between mechanical properties and biological properties in porous structure design, enhancing the mechanical properties of the structure often leads to a decrease in biological properties. In such a trade-off, simple porous structural units are difficult to maintain a good balance between biological and mechanical properties like complex porous structures in nature.

[0005] Therefore, how to design a porous structure that can express complex shapes and regulate it is a difficult problem. At the same time, for complex porous structure units, if traditional STL files are used to represent them, a large number of triangular facets are required to represent the complex geometric surface of the porous structure, which will result in a large memory consumption. Summary of the invention

[0006] The purpose of the present invention is to address the deficiencies of the prior art and propose a porous structure design method based on implicit B-splines, which solves the problems of large storage memory consumption caused by designing porous structures with complex geometric surfaces using a large number of triangular facets and difficulty in regulating the porous structure through a large number of triangular facets.

[0007] The technical solution adopted by the present invention is:

[0008] 1) Input the porous structure discrete distance field file of the real porous structure, which contains the following information: the number of grids along three directions in the discrete distance field and the position distance field value of each grid;

[0009] 2) Selecting a three-variable implicit B-spline function to iteratively fit the position distance field values ​​of each grid in the discrete distance field, and obtaining a continuous distance distribution field C(u, v, w) after iterative fitting;

[0010] 3) According to the structural type of the porous structure input by the user, the porous structure in space is obtained by combining the continuous distance distribution field C(u, v, w);

[0011] Among them, the isosurface threshold fields t0(u, v, w), t1(u, v, w), and t2(u, v, w) are all obtained by expressing the three-variable B-spline function.

[0012] 4) The porous structure is stored in ITDF file format, and the porous structure is further edited and input into a 3D printer for production.

[0013] The porous structure discrete distance field file is obtained by scanning a real porous structure into a point cloud and then converting it.

[0014] The porous structure is a structure with abundant pores inside, such as a porous bone scaffold for bone tissue engineering, a porous catalyst carrier for chemical reactions, and a honeycomb-like structure for material lightweighting.

[0015] The real porous structure is usually bone tissue, zeolite, sponge and the like.

[0016] The discrete distance field is divided into multiple grids, and the center of each grid has a position distance field value; the position distance field value varies with the distance from the solid part of the porous structure, and the closer to the solid part of the porous structure, the larger / smaller the position distance field value, and the solid part refers to the material part that does not contain voids in the porous structure.

[0017] The discrete distance field is a discrete field representing the shape of the porous structure, which is specifically expressed as a distance field value of a position in space from a solid part of the porous structure.

[0018] In step 2), the three-variable implicit B-spline function is selected to fit the discrete distance field data points, which is specifically manifested as follows:

[0019] The discrete distance field is established in a three-dimensional Euclidean space, and the discrete distance field D is expressed as D = {x α,β,γ |1≤α≤w, 1≤β≤h, 1≤γ≤d}, α, β, γ represent the grid indexes in the three directions of the three-dimensional Euclidean space, w, h, d represent the total number of grids in the three directions of the three-dimensional Euclidean space; each grid of the discrete distance field D is evenly distributed in the unit cube of the three-dimensional Euclidean space, the corner points of the unit cube in the first quadrant coincide with the origin of the three-dimensional Euclidean space, and the three sides of the unit cube fall on the coordinate axes of the three-dimensional Euclidean space. For each grid center position (u α , v β , w γ ) is set to be close enough to the three-variable implicit B-spline function, and the following objective function is established:

[0020]

[0021]

[0022]

[0023] Among them, C(u α , v β , w γ ) represents the center position of the grid (u α , v β , w γ ) is the value of the continuous distance distribution field at x α,β,γ Indicates the center position of the grid (u α , v β , w γ ), i, j, k represent the control coefficients in three directions respectively. coordinate Ordinal, c ijk is the (i, j, k)th control coefficient of the three-variable implicit B-spline function, n u , n v , n w is the number of control points of the three-variable implicit B-spline function in three directions, B i,j,k (u α , v β , w γ ) is the (i, j, k)th basis function of the three-variable implicit B-spline function at position (u α , v β , w γ ), L represents the overall fitting error; u α , v β , w γ They represent the coordinates of the center position of the grid in three directions respectively, and C(u, v, w) represents the value calculated by substituting the spatial position (u, v, w) into the three-variable implicit B-spline function;

[0024] Taking the minimum objective function as the goal, each control coefficient c is obtained by iteratively fitting and solving the above objective function. ijk , thereby determining the continuous distance distribution field C(u, v, w).

[0025] In the specific implementation, the following method is used for iterative fitting and solving:

[0026] The position distance field values ​​at the center of the grid are uniformly selected at equal intervals, and the selected position distance field values ​​are used as the initial control coefficients of the continuous distance distribution field C(u, v, w). The three-variable implicit B-spline function is iteratively fitted according to the following equation. Each iteration generates a continuous distance distribution field, and the continuous distance distribution field C of the l-1th iteration is (l-1) (u, v, w) generates the continuous distance distribution field C of the lth iteration (l) (u, v, w):

[0027]

[0028]

[0029]

[0030]

[0031] B i,j,k (u α , v β , w γ )≠0

[0032] Where: l is the number of iterations, i, j, k represent the coordinate ordinal numbers of the control coefficients in three directions, I i,j,k Indicates that B i,j,k (u α , v β , w γ )≠0, represents the difference vector of the l-1th iteration, Indicates that at position (u α , v β , w γ ) is generated by the three-variable implicit B-spline function at the l-1th iteration and the value of the grid center position (u α , v β , w γ ) is the distance field value x α,β,γ The error value, B i,j,k (u α , v β , w γ ) represents the (i, j, k)th basis function of the three-variable implicit B-spline function at the center of the grid (u α , v β , w γ ) value; and represents the (i, j, k)th control coefficient of the continuous distance distribution field generated after the lth and l-1th iterations, respectively, C (l) (u, v, w) represents the continuous distance distribution field generated by the lth iteration, B i,j,k (u, v, w) is the value of the (i, j, k)th three-variable B-spline basis function at (u, v, w);

[0033] After each iteration, the overall fitting error ∈ l , the overall fitting error ∈ of the last two times is calculated in real time l The difference is compared with the expected fitting error threshold ∈ to terminate the iteration.

[0034] After the lth iteration, the overall fitting error ∈ of all grid points is calculated based on the error value obtained in the lth iteration l :

[0035]

[0036]

[0037] In the formula, C (l) (u α , v β , w γ ) indicates that the coordinates are (u α , v β , w γ ) is the value of the continuous distance distribution field generated by the grid center point at the lth iteration;

[0038] The overall fitting error ∈ l The overall fitting error ∈ with the previous iteration l-1 Compare and subtract, and judge:

[0039] When the difference is less than the expected fitting error threshold ∈, the iteration terminates; otherwise, the next iteration is performed until the difference between the overall fitting error of the current iteration and the overall fitting error of the previous iteration is less than the expected fitting error threshold ∈.

[0040] In the step 3), if the porous structure is represented as a body type, a first isosurface threshold field t1(u, v, w) and a second isosurface threshold field t2(u, v, w) are set according to the expected data input by the user, and then t1(u, v, w)≤C(u, v, w)≤t2(u, v, w) is used as a condition, and the area formed by all positions (u, v, w) that meet the condition is solved to obtain a porous structure of a solid in space;

[0041] In the step 3), if the porous structure is represented as a surface type, the isosurface threshold field t0(u, v, w) is set according to the expected data input by the user, and then the area formed by all positions (u, v, w) that meet the condition C(u, v, w) = t0(u, v, w) is solved to obtain the porous structure of the curved surface in space.

[0042] In step 3), the first isosurface threshold field t1(u, v, w) is obtained in the following manner:

[0043] The first isosurface threshold field is established in a three-dimensional Euclidean space. In order to obtain the first isosurface threshold field, the user needs to input a discrete first isosurface threshold field. The discrete first isosurface threshold field is established in a three-dimensional Euclidean space. The discrete first isosurface threshold field is expressed as α1, β1, γ1 represent the grid indexes in three directions in the three-dimensional Euclidean space, w1, h1, d1 represent the total number of grids in three directions in the three-dimensional Euclidean space, represents the first isosurface threshold field value expected by the user at the (α1, β1, γ1)th grid position;

[0044] Each grid of the discrete first isosurface threshold field is evenly distributed in the unit cube of the three-dimensional Euclidean space. The corner points of the unit cube in the first quadrant coincide with the origin of the three-dimensional Euclidean space, and the three sides of the unit cube fall on the coordinate axes of the three-dimensional Euclidean space. The center position of each grid in the three-dimensional Euclidean space is set to be close enough to the three-variable B-spline function, and the following objective function is established:

[0045]

[0046]

[0047]

[0048] in, Indicates the center position of the grid The value of the first isosurface threshold field at , Indicates the center position of the grid The discrete first isosurface threshold field value at , i, j, k represent the control coefficient in three directions respectively. coordinate Ordinal numbers, is the (i, j, k)th control coefficient of the three-variable B-spline function, is the number of control points of the three-variable B-spline function in three directions, is the (i, j, k)th basis function of the three-variable B-spline function at position The value at , L1 represents the overall fitting error; They represent the coordinates of the center position of the grid in three directions respectively, and t1(u, v, w) represents the value calculated by substituting the spatial position (u, v, w) into the three-variable B-spline function;

[0049] Taking the minimum objective function as the goal, each control coefficient is obtained by iteratively fitting and solving the above objective function. Thus, the first isosurface threshold field t1(u, v, w) is determined;

[0050] In step 3), the second isosurface threshold field t2(u, v, w) is obtained in the following manner:

[0051] The second isosurface threshold field is established in a three-dimensional Euclidean space. In order to obtain the second isosurface threshold field, the user needs to input a discrete second isosurface threshold field. The discrete second isosurface threshold field is established in a three-dimensional Euclidean space. The discrete second isosurface threshold field is expressed as α2, β2, γ2 represent the grid indexes in three directions in the three-dimensional Euclidean space, w2, h2, d2 represent the total number of grids in three directions in the three-dimensional Euclidean space, Represents the second isosurface threshold field value expected by the user at the (α2, β2, γ2)th grid position.

[0052] Each grid of the discrete second isosurface threshold field is evenly distributed in the unit cube of the three-dimensional Euclidean space. The corner points of the unit cube in the first quadrant coincide with the origin of the three-dimensional Euclidean space, and the three sides of the unit cube fall on the coordinate axes of the three-dimensional Euclidean space. The center position of each grid in the three-dimensional Euclidean space is set to be close enough to the three-variable B-spline function, and the following objective function is established:

[0053]

[0054]

[0055]

[0056] in, Indicates the center position of the grid The value of the second isosurface threshold field at , Indicates the center position of the grid The discrete second isosurface threshold field value at , i, j, k represent the control coefficient in three directions respectively. coordinate Ordinal numbers, is the (i, j, k)th control coefficient of the three-variable B-spline function, is the number of control points of the three-variable B-spline function in three directions, is the (i, j, k)th basis function of the three-variable B-spline function at position The value at , L2 represents the overall fitting error; They represent the coordinates of the center position of the grid in three directions respectively, and t2(u, v, w) represents the value calculated by substituting the spatial position (u, v, w) into the three-variable B-spline function;

[0057] Taking the minimum objective function as the goal, each control coefficient is obtained by iteratively fitting and solving the above objective function. Thus, the second isosurface threshold field t2 (u, v, w) is determined.

[0058] In step 3), the isosurface threshold field t0(u, w, v) is obtained in the following manner:

[0059] The zeroth isosurface threshold field is established in a three-dimensional Euclidean space. In order to obtain the zeroth isosurface threshold field, the user needs to input a discrete zeroth isosurface threshold field. The discrete zeroth isosurface threshold field is established in a three-dimensional Euclidean space. The discrete zeroth isosurface threshold field is expressed as α0, β0, γ0 represent the grid indexes in three directions in the three-dimensional Euclidean space, w0, h0, d0 represent the total number of grids in three directions in the three-dimensional Euclidean space, Represents the zeroth isosurface threshold field value expected by the user at the (α0, β0, γ0)th grid position.

[0060] Each grid of the discrete zeroth isosurface threshold field is evenly distributed in the unit cube of the three-dimensional Euclidean space. The corner points of the unit cube in the first quadrant coincide with the origin of the three-dimensional Euclidean space, and the three sides of the unit cube fall on the coordinate axes of the three-dimensional Euclidean space. The center position of each grid in the three-dimensional Euclidean space is set to be close enough to the three-variable B-spline function, and the following objective function is established:

[0061]

[0062]

[0063]

[0064] in, Indicates the center position of the grid The value of the zeroth isosurface threshold field at Indicates the center position of the grid The discrete zeroth isosurface threshold field value at , i, j, k represent the control coefficient in three directions respectively. coordinate Ordinal numbers, is the (i, j, k)th control coefficient of the three-variable B-spline function, is the number of control points of the three-variable B-spline function in three directions, is the (i, j, k)th basis function of the three-variable B-spline function at position The value at , L0 represents the overall fitting error; They represent the coordinates of the center position of the grid in three directions respectively, and t0(u, v, w) represents the value calculated by substituting the spatial position (u, v, w) into the three-variable B-spline function;

[0065] Taking the minimum objective function as the goal, each control coefficient is obtained by iteratively fitting and solving the above objective function. Thus, the zeroth isosurface threshold field t0(u, v, w) is determined.

[0066] After step 3), the obtained porous structure is further subjected to manufacturing inspection to determine whether the porous structure meets the actual requirements: if not, if the porous structure is a body type, the first isosurface threshold field or the second isosurface threshold field or the continuous distance distribution field is locally regulated; if the porous structure is a surface type, the zeroth isosurface threshold field or the continuous distance distribution field is locally regulated, otherwise no local regulation is performed.

[0067] The local regulation is to adjust the expected isosurface threshold value or continuous distance value at the local position point in the isosurface threshold field and the continuous distance distribution field, and then regenerate the isosurface threshold field or the continuous distance distribution field in the same manner as in step 2) and step 3), adjust the shape and wall thickness of the porous structure, and then adjust the porosity and pore size of the porous structure.

[0068] In the step 4), the information stored in the ITDF file format includes the following:

[0069] The structural type of porous structure: the surface type porous structure is FACET, and the body type porous structure is SOILD;

[0070] The number of control coefficients of the three-variable implicit B-spline function of the continuous distance distribution field in three directions;

[0071] Control coefficients of three-variable implicit B-spline function;

[0072] Node vector information of the continuous distance distribution field in three directions.

[0073] The knot vector is the set of non-decreasing numbers required by the three-variable B-spline function.

[0074] For a porous structure represented by a surface, only one isosurface threshold field is recorded, while for a porous structure represented by a volume, two isosurface threshold fields are recorded. The isosurface threshold field part records the number of control points of the isosurface threshold field in three directions, the control point values ​​of the isosurface threshold field, and the node vector information of the isosurface threshold field in three directions.

[0075] In this case, the information stored in the present invention does not need to save the information of each vertex and surface of the triangular mesh, but only the information of a solid structure or a surface structure, which greatly reduces the amount of stored data.

[0076] The present invention proposes a method for representing porous structure by implicit B-spline, and based on the method, a surface porous structure model and a volume porous structure model are generated.

[0077] Under the scheme of the present invention, the porous structure can be represented, generated, and edited using implicit B-splines. Under this representation, the internal smoothness of the porous structure is guaranteed. At the same time, by adjusting t0(u, v, w) in the surface form and t1(u, v, w) and t2(u, v, w) in the volume form, i.e., the isosurface threshold field, the porosity of the porous structure can be effectively controlled, ensuring the controllability of the porous structure; by adjusting the continuous distance distribution field C(u, v, w), the shape of the porous structure can be effectively controlled.

[0078] Compared with the porous structure stored in the traditional STL file format, the porous structure represented by the implicit B-spline in the present invention can significantly reduce the storage space.

[0079] Compared with the prior art, the present invention has the following advantages:

[0080] 1. This method uses implicit B-splines to represent porous structures, and uses isosurface threshold fields and continuous distance distribution fields to represent porous structures. Due to the easy-to-adjust characteristics of B-splines, the isosurface threshold fields and continuous distance distribution fields can be easily adjusted, and it also has the ability to regulate the mechanical and biological properties of the final structure.

[0081] The adjustment of the continuous distance distribution field can locally control the shape of the porous structure and enrich the geometric details of the porous structure; the adjustment of the isosurface threshold field can adjust the wall thickness of the corresponding position of the porous structure to regulate the porosity and pore size of the porous structure.

[0082] 2. Due to the rich expressive power of implicit functions, porous structures with complex details can be accurately described, and the smoothness of implicit B-splines ensures the smoothness of the final porous structure.

[0083] 3. The use of ITDF format to store porous structures represented by implicit B-spline greatly reduces file memory compared to traditional STL files for storing porous structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0085] The present invention will be further described below in conjunction with the accompanying drawings and specific implementations.

[0086] like Figure 1 As shown, the embodiments of the present invention and their implementation process are as follows:

[0087] 1) Introduction of discrete distance field

[0088] Input the discrete distance field to represent the porous structure.

[0089] In the specific implementation, the following discrete distance fields are included as examples:

[0090] A) For porous structures represented by voxels, one voxel is used as a grid. The porous structure is represented by two phases, 0 and 1. Phase 0 represents holes, and phase 1 represents material.

[0091] The discrete distance field of the porous structure is represented by distance fields such as Manhattan distance field (the distance field value corresponding to each voxel in the space is the distance from the voxel to the nearest voxel with a different phase), KNN density field (the distance field value corresponding to each voxel in the space is the density of voxels with a phase of 1 in a certain neighborhood of the voxel), and DTM (distance-to-measure) distance field (the distance field value corresponding to each voxel in the space is the average distance from the voxel to the nearest few voxels with a phase of 1).

[0092] The above distance fields all satisfy the characteristic of containing information about the shape of the porous structure. Other distance field values ​​that satisfy the distance voxels in space representing the distance to the physical part of the porous structure can also be used as input discrete distance fields of this method.

[0093] B) For a porous structure represented by a grid, its bounding box in the three-dimensional Euclidean space is divided into a uniform grid. The discrete distance field is calculated by taking a grid data point at the center of each grid unit and calculating the distance field value of each point. For example, when calculating a signed distance field (the distance from a grid point to the nearest triangle, which is positive on one side of the normal and negative on the other side) or an unsigned distance field (the distance from a grid point to the nearest triangle), the distance field value of each grid point is used as the discrete distance field.

[0094] The discrete fields of other grid porous structures that satisfy the distance field values ​​of positions in space that are close or far from the grid representing the physical part of the porous structure can also be used as the input discrete distance fields of this method.

[0095] C) For the porous structure represented by the implicit function, the bounding box of the limited area in the three-dimensional Euclidean space is divided into a uniform grid, and the grid data point is taken at the center of each grid unit.

[0096] Taking the three-dimensional periodic minimal surface as an example, the porous structure is represented by the implicit function φ(x, y, z) = c, where φ(x, y, z) is generally a trigonometric function. The bounding box of the porous structure to be represented can be given in advance, and grid data points are uniformly taken inside the bounding box. The value φ(x0, y0, z0)-c at each point (x0, y0, z0) is calculated, where c is represented as the isosurface threshold of the three-dimensional periodic minimal surface, and the value calculated at each grid point is used as the value of the discrete distance field.

[0097] Other implicit function porous structure discrete fields that satisfy the distance implicit function in space to represent the distance field value of the position of the physical part of the porous structure can also be used as the input discrete distance field of this method.

[0098] D) For the porous structure represented by the point cloud, its bounding box in the three-dimensional Euclidean space is divided into a uniform grid, and the grid data point is taken at the center of each grid unit.

[0099] The calculation of the distance field takes the k-average distance field as an example. For each grid point, the average value of the distance to the nearest k points in the point cloud is calculated as the data value at the grid point, and the distance value of the grid point is used as the discrete distance field.

[0100] The point cloud porous structure discrete field that satisfies the distance field value of other positions in space that are far or near from the point cloud representing the entity part of the porous structure can also be used as the input discrete distance field of this method.

[0101] 2) Constructing a continuous distance distribution field

[0102] The discrete distance field is established in a three-dimensional Euclidean space, and the discrete distance field D is expressed as D = {x α,β,γ |1≤α≤w, 1≤β≤h, 1≤γ≤d, α, β, γ represent the grid indexes in three directions in the three-dimensional Euclidean space, w, h, d represent the total number of grids in three directions in the three-dimensional Euclidean space;

[0103] Each grid of the discrete distance field D is evenly distributed in the unit cube of the three-dimensional Euclidean space. The corner points of the unit cube in the first quadrant coincide with the origin of the three-dimensional Euclidean space, and the three sides of the unit cube fall on the coordinate axes of the three-dimensional Euclidean space. For each grid center position (u α , v β , w γ ) is set to be close enough to the three-variable implicit B-spline function, and the following objective function is established:

[0104]

[0105]

[0106]

[0107] Among them, C(u α , v β , w γ ) represents the center position of the grid (u α , v β , w γ ) is the value of the continuous distance distribution field at x α,β,γ Indicates the center position of the grid (u α , v β , w γ ), i, j, k represent the control coefficients in three directions respectively. coordinate Ordinal, cijk is the (i, j, k)th control coefficient of the three-variable implicit B-spline function, n u , n v , n w is the number of control points of the three-variable implicit B-spline function in three directions, B i,j,k (u α , v β , w γ ) is the (i, j, k)th basis function of the three-variable implicit B-spline function at position (u α , v β , w γ ), L represents the overall fitting error; u α , v β , w γ They represent the coordinates of the center position of the grid in three directions respectively, and C(u, v, w) represents the value calculated by substituting the spatial position (u, v, w) into the three-variable implicit B-spline function;

[0108] Taking the minimum objective function as the goal, each control coefficient c is obtained by iteratively fitting and solving the above objective function. ijk , thereby determining the continuous distance distribution field C(u, v, w).

[0109] In the specific implementation, the following method is used for iterative fitting and solving:

[0110] The position distance field values ​​at the center of the grid are uniformly selected at equal intervals, and the selected position distance field values ​​are used as the initial control coefficients of the continuous distance distribution field C(u, v, w). The three-variable implicit B-spline function is iteratively fitted according to the following equation. Each iteration generates a continuous distance distribution field, and the continuous distance distribution field C of the l-1th iteration is (l-1) (u, v, w) generates the continuous distance distribution field C of the lth iteration (l) (u, v, w):

[0111]

[0112]

[0113]

[0114]

[0115] B i,j,k (u α , v β , w γ )≠0

[0116] Where: l is the number of iterations, i, j, k represent the control coefficients in three directions respectively. coordinate Ordinal number, I i,j,k Indicates that B i,j,k (u α , v β , w γ )≠0, represents the difference vector of the l-1th iteration, Indicates that at position (u α , v β , w γ ) is generated by the three-variable implicit B-spline function at the l-1th iteration and the value of the grid center position (u α , v β , w γ ) is the distance field value x α,β,γ The error value, B i,j,k (u α , v β , w γ ) represents the (i, j, k)th basis function of the three-variable implicit B-spline function at the center of the grid (u α , v β , w γ ) value; and represents the (i, j, k)th control coefficient of the continuous distance distribution field generated after the lth and l-1th iterations, respectively, C (l) (u, v, w) represents the continuous distance distribution field generated by the lth iteration, B i,j,k (u, v, w) is the value of the (i, j, k)th three-variable B-spline basis function at (u, v, w);

[0117] After each iteration, the overall fitting error ∈ l , the overall fitting error ∈ of the last two times is calculated in real time l The difference is compared with the expected fitting error threshold ∈ to terminate the iteration.

[0118] After the lth iteration, the overall fitting error ∈ of all grid points is calculated based on the error value obtained in the lth iteration l :

[0119]

[0120]

[0121] In the formula, C (l) (u α , v β , w γ ) indicates that the coordinates are (u α, v β , w γ ) is the value of the continuous distance distribution field generated by the grid center point at the lth iteration;

[0122] The overall fitting error ∈ l The overall fitting error ∈ with the previous iteration l-1 Compare and subtract, and judge:

[0123] When the difference is less than the expected fitting error threshold ∈, the iteration terminates; otherwise, the next iteration is performed until the difference between the overall fitting error of the current iteration and the overall fitting error of the previous iteration is less than the expected fitting error threshold ∈.

[0124] 3) Representation of porous structure

[0125] If the porous structure is represented as a body type, the first isosurface threshold field t1(u, v, w) and the second isosurface threshold field t2(u, v, w) are set according to the expected data input by the user, and then the area formed by all positions (u, v, w) that meet the condition of t1(u, v, w)≤C(u, v, w)≤t2(u, v, w) is solved to obtain the porous structure entity in space;

[0126] If the porous structure is represented as a surface type, the isosurface threshold field t0(u, w, v) is set according to the expected data input by the user, and then the area formed by all positions (u, v, w) that meet the condition is solved with C(u, v, w) = t0(u, w, v) as the condition to obtain the porous structure surface in space;

[0127] For porous structures represented by surfaces, an isosurface threshold field t0(u, v, w) needs to be designed; for porous structures represented by volumes, two isosurface threshold fields t1(u, v, w) and t2(u, v, w) need to be designed respectively.

[0128] 4) Editing and adjusting porous structures

[0129] Based on the feedback from the porous structure manufacturing results, the porous structure is locally regulated.

[0130] The main ways of regulation are regulation of the isosurface threshold field and regulation of the continuous distance distribution field. The continuous distance distribution field can regulate the shape of the porous structure, and the regulation of the isosurface threshold field can regulate the shape and wall thickness of the porous structure, thereby regulating the porosity, pore size and other related properties of the porous structure.

[0131] Local regulation is to adjust the isosurface threshold value or continuous distance value expected at the local position point in the isosurface threshold field and the continuous distance distribution field, and then regenerate the isosurface threshold field or the continuous distance distribution field in the same way as step 2) and step 3), specifically to adjust the shape and wall thickness of the porous structure, and then adjust the porosity and pore size of the porous structure.

[0132] 5) Store in ITDF file format

[0133] The porous structure represented by the implicit B-spline is stored in the ITDF file format and input into a 3D printer for printing to obtain the final model product.

[0134] Stored in ITDF file format, the information stored in the file includes the following:

[0135] The structural type of porous structure: the surface type porous structure is FACET, and the body type porous structure is SOILD;

[0136] The number of control coefficients of the three-variable implicit B-spline function of the continuous distance distribution field in three directions;

[0137] Control coefficients of three-variable implicit B-spline function;

[0138] Node vector information of the continuous distance distribution field in three directions.

[0139] For a porous structure represented by a surface, only one isosurface threshold field needs to be recorded, while for a porous structure represented by a volume, two isosurface threshold fields need to be recorded. The isosurface threshold field part needs to record the number of control points of the isosurface threshold field in three directions; the control point value of the isosurface threshold field; and the node vector information of the isosurface threshold field in three directions.

[0140] The knot vector is a set of non-decreasing numbers required for the definition of the three-variable B-spline function.

[0141] Therefore, the present invention realizes a porous structure design method based on implicit B-splines; based on implicit B-splines, complex porous structure information is preserved, and two types of implicit porous structures are provided; the spatial information of the porous structure is expressed by using a continuous distance distribution field; the function of interactively modifying the porous structure is provided by using an isosurface threshold field; the representability and controllability of complex porous structures are greatly improved; and an ITDF file format is provided to preserve the porous structure, which greatly reduces the memory requirement for storing complex porous structures.

Claims

1. A porous structure design method based on implicit B-splines, characterized in that: The method includes the following steps: 1) Input the porous structure discrete distance field file of the real porous structure, including: the number of grids along three directions in the discrete distance field and the position distance field value of each grid; 2) Select the three-variable implicit B-spline function to iteratively fit the position distance field values ​​of each grid in the discrete distance field to obtain the continuous distance distribution field C(u,v,w); 3) According to the structural type of the porous structure input by the user, the porous structure in space is obtained by combining the continuous distance distribution field C(u, v, w); In the step 3), if the porous structure is a body type, a first isosurface threshold field t1(u, v, w) and a second isosurface threshold field t2(u, v, w) are set according to the expected data input by the user, and then the area formed by all positions (u, v, w) that meet the condition of t1(u, v, w)≤C(u, v, w)≤t2(u, v, w) is solved to obtain a porous structure of a solid in space; In the step 3), if the porous structure is represented as a surface type, an isosurface threshold field t0(u,w,v) is set according to the expected data input by the user, and then the area formed by all positions (u,v,w) that meet the condition is solved with C(u,v,w)=t0(u,w,v) as the condition to obtain the porous structure of the curved surface in space; In step 3), the first isosurface threshold field t1(u,v,w) is obtained in the following manner: The first isosurface threshold field is established in a three-dimensional Euclidean space. In order to obtain the first isosurface threshold field, the user needs to input a discrete first isosurface threshold field. The discrete first isosurface threshold field is established in a three-dimensional Euclidean space. The discrete first isosurface threshold field is expressed as α1, β1, γ1 represent the grid indexes in three directions in the three-dimensional Euclidean space, w1, h1, d1 represent the total number of grids in three directions in the three-dimensional Euclidean space, represents the first isosurface threshold field value expected by the user at the (α1, β1, γ1)th grid position; Each grid of the discrete first isosurface threshold field is evenly distributed in the three-dimensional Euclidean space. The center position of each grid in the three-dimensional Euclidean space is set to be close enough to the three-variable B-spline function, and the following objective function is established: in, Indicates the center position of the grid The value of the first isosurface threshold field at , Indicates the center position of the grid The discrete first isosurface threshold field value at , i, j, k represent the coordinate ordinal of the control coefficient in three directions, is the (i, j, k)th control coefficient of the three-variable B-spline function, is the number of control points of the three-variable B-spline function in three directions, is the (i,j,k)th basis function of the three-variable B-spline function at position The value at , L1 represents the overall fitting error; They represent the coordinates of the center position of the grid in three directions respectively, and t1(u,v,w) represents the value calculated by substituting the spatial position (u,v,w) into the three-variable B-spline function; Taking the minimum objective function as the goal, each control coefficient is obtained by iteratively fitting and solving the above objective function. Thus, the first isosurface threshold field t1(u,v,w) is determined; In step 3), the second isosurface threshold field t2(u,v,w) is obtained in the following manner: The second isosurface threshold field is established in a three-dimensional Euclidean space. In order to obtain the second isosurface threshold field, the user needs to input a discrete second isosurface threshold field. The discrete second isosurface threshold field is established in a three-dimensional Euclidean space. The discrete second isosurface threshold field is expressed as α2, β2, γ2 represent the grid indexes in three directions in the three-dimensional Euclidean space, w2, h2, d2 represent the total number of grids in three directions in the three-dimensional Euclidean space, represents the second isosurface threshold field value expected by the user at the (α2, β2, γ2)th grid position; Each grid of the discrete second isosurface threshold field is evenly distributed in the three-dimensional Euclidean space. The center position of each grid in the three-dimensional Euclidean space is set to be close enough to the three-variable B-spline function, and the following objective function is established: in, Indicates the center position of the grid The value of the second isosurface threshold field at , Indicates the center position of the grid The discrete second isosurface threshold field value at , I, J, k represent the coordinate ordinal of the control coefficient in three directions, is the (i, j, K)th control coefficient of the three-variable B-spline function, is the number of control points of the three-variable B-spline function in three directions, is the (I,j,k)th basis function of the three-variable B-spline function at position The value at , L2 represents the overall fitting error; They represent the coordinates of the center of the grid in three directions, and t2(u,v,w) represents the value calculated by substituting the spatial position (u,v,w) into the three-variable B-spline function; Taking the minimum objective function as the goal, each control coefficient is obtained by iteratively fitting and solving the above objective function. Thus, the second isosurface threshold field t2(u,v,w) is determined; In step 3), the isosurface threshold field t0(u,w,v) is obtained in the following manner: The zeroth isosurface threshold field is established in three-dimensional Euclidean space. In order to obtain the zeroth isosurface threshold field, the user needs to input the discrete zeroth isosurface threshold field. The discrete zeroth isosurface threshold field is established in three-dimensional Euclidean space. The discrete zeroth isosurface threshold field is expressed as α0, β0, γ0 represent the grid indexes in three directions in the three-dimensional Euclidean space, w0, h0, d0 represent the total number of grids in three directions in the three-dimensional Euclidean space, represents the zeroth isosurface threshold field value expected by the user at the (α0, β0, γ0)th grid position; Each grid of the discrete zeroth isosurface threshold field is evenly distributed in the three-dimensional Euclidean space. The center position of each grid in the three-dimensional Euclidean space is set to be close enough to the three-variable B-spline function, and the following objective function is established: in, Indicates the center position of the grid The value of the zeroth isosurface threshold field at Indicates the center position of the grid The discrete zeroth isosurface threshold field value at , i, j, k represent the coordinate ordinal numbers of the control coefficients in three directions, is the (i, j, k)th control coefficient of the three-variable B-spline function, is the number of control points of the three-variable B-spline function in three directions, is the (i,j,k)th basis function of the three-variable B-spline function at position The value at , L0 represents the overall fitting error; They represent the coordinates of the center of the grid in three directions, and t0(u,v,w) represents the value calculated by substituting the spatial position (u,v,w) into the three-variable B-spline function; Taking the minimum objective function as the goal, each control coefficient is obtained by iteratively fitting and solving the above objective function. Thus, the zeroth isosurface threshold field t0(u,v,w) is determined; 4) Store the porous structure in ITDF file format.

2. The porous structure design method based on implicit B-splines according to claim 1, characterized in that: The porous structure is a porous bone scaffold for bone tissue engineering, a porous catalyst carrier for chemical reactions, and a honeycomb-like structure for lightweighting materials.

3. The porous structure design method based on implicit B-splines according to claim 1, characterized in that: The discrete distance field is divided into multiple grids, and the center of each grid has a position distance field value; the position distance field value varies with the distance from the solid part of the porous structure, and the closer to the solid part of the porous structure, the larger / smaller the position distance field value, and the solid part refers to the material part that does not contain voids in the porous structure.

4. The method for designing porous structures based on implicit B-splines according to claim 1, characterized in that: In step 2), the three-variable implicit B-spline function is selected to fit the discrete distance field data points, which is specifically manifested as follows: The discrete distance field is established in a three-dimensional Euclidean space, and the discrete distance field D is expressed as D = {x α,β,γ |1≤α≤w, 1≤β≤h, 1≤γ≤d}, α, β, γ represent the grid indexes in three directions in the three-dimensional Euclidean space, w, h, d represent the total number of grids in three directions in the three-dimensional Euclidean space; Each grid of the discrete distance field D is uniformly distributed in the three-dimensional Euclidean space. For each grid center position (u α ,v β ,w γ ) is set to be close enough to the three-variable implicit B-spline function, and the following objective function is established: Among them, C(u α ,b β ,w γ ) represents the center position of the grid (u α ,v β ,w γ ) is the value of the continuous distance distribution field at x α,β,γ Indicates the center position of the grid (u α ,v β ,w γ ), i, j, k represent the control coefficients in three directions respectively. coordinate Ordinal, c ijk is the (i, j, k)th control coefficient of the three-variable implicit B-spline function, n u ,n v ,n w is the number of control points of the three-variable implicit B-spline function in three directions, B i,j,k (u α ,v β ,w γ ) is the (i,j,k)th basis function of the three-variable implicit B-spline function at position (u α ,v β ,w γ ), L represents the overall fitting error; u α ,v β ,w γ They represent the coordinates of the center position of the grid in three directions respectively, and C(u,v,w) represents the value calculated by substituting the spatial position (u,v,w) into the three-variable implicit B-spline function; Taking the minimum objective function as the goal, each control coefficient c is obtained by iteratively fitting and solving the above objective function. ijk , thereby determining the continuous distance distribution field C(u,v,w).

5. The porous structure design method based on implicit B-splines according to claim 1, characterized in that: After step 3), the obtained porous structure is further subjected to manufacturing inspection to determine whether the porous structure meets the actual requirements: if not, if the porous structure is a body type, the first isosurface threshold field or the second isosurface threshold field or the continuous distance distribution field is locally regulated; if the porous structure is a surface type, the zeroth isosurface threshold field or the continuous distance distribution field is locally regulated.

6. The method for designing porous structures based on implicit B-splines according to claim 5, characterized in that: The local regulation is to obtain the expected isosurface threshold value or continuous distance value at the local position point in the isosurface threshold field and the continuous distance distribution field, and then regenerate the isosurface threshold field or the continuous distance distribution field in the same way as in step 2).

7. The method for designing porous structures based on implicit B-splines according to claim 1, characterized in that: In the step 4), the information stored in the ITDF file format includes the following: The structural type of porous structure: the surface type porous structure is FACET, and the body type porous structure is SOILD; The number of control coefficients of the three-variable implicit B-spline function of the continuous distance distribution field in three directions; Control coefficients of three-variable implicit B-spline function; Node vector information of the continuous distance distribution field in three directions.