Intelligent slicing optimization method and system for 3D printer

Abstract

The present invention relates to the technical field of 3D printing slicing, and discloses an intelligent slicing optimization method and system for a 3D printer, the method comprising the following steps: S1: acquiring a 3D model to be sliced, and selecting feature points on the surface of the 3D model; S2: extracting a feature index of each feature point, and calculating the precision value of each feature point based on the feature index; S3: dividing the 3D model into sub-models based on the precision value of each feature point; S4: setting a slicing plan for each sub-model based on the precision value of each feature point; S5: simulating slicing of each sub-model based on the slicing plan, adjusting the slicing plan based on the simulated slicing, and generating a slicing scheme for each sub-model; the present invention can automatically identify model features and adjust slicing parameters accordingly, effectively improve printing precision, reduce printing time, and reduce printing failure rate.

Description

Technical Field

[0001] The present invention relates to the technical field of 3D printing slicing, and in particular to an intelligent slicing optimization method and system for a 3D printer. Background Art

[0002] In recent years, 3D printing technology has developed rapidly. 3D printing technology can reduce the cost of product development cycle and improve the customization level of products. Its application scope covers a variety of fields such as aerospace, automobile manufacturing, medical equipment, and construction industry. The core of 3D printing lies in the process of converting digital models into physical objects. This process usually involves slicing the model layer by layer, and then stacking materials layer by layer according to the slicing information to build the final product. In order to ensure the quality of printing, 3D printing software usually provides some basic slicing setting options, such as layer thickness, filling density, etc. With the increasing application of 3D printing technology, the traditional one-size-fits-all slicing method has been difficult to meet the high-precision requirements in different application scenarios. For example, in the 3D printing of the jewelry industry, especially when dealing with fine patterns and complex inlay structures, how to ensure printing quality and efficiency is an urgent problem for manufacturers.

[0003] Most current 3D printing technologies use a uniform layer thickness for slicing. For areas with rich surface details or large curvature, using a uniform layer thickness will cause distortion in the printing results and loss of details. For areas with smooth or simple surfaces, a uniform layer thickness will increase printing time. In the prior art, feature recognition of 3D models often relies on manually set rules or rough statistical analysis, and lacks a refined analysis of the surface features of the model. This makes it difficult to accurately capture the key features of the model when processing models with complex geometric shapes or surface textures, which in turn affects the printing quality and makes the printing quality of different areas vary significantly. Traditional model slicing methods are often based on mesh division and analyze the vertices of the triangular mesh. They cannot ensure the uniform distribution of feature points and may lose detailed features of the model surface.

[0004] For example, a Chinese patent with authorization announcement number CN112069725B discloses a method and device for obtaining high-precision slices for a 3D printer, which belongs to the technical field of 3D printers. The method includes: obtaining a set of sampling points; using a polygon approximation algorithm to obtain a closed polygon corresponding to the set of sampling points; inputting a data sequence corresponding to the closed polygon into a pre-trained contour generation model to obtain an object contour corresponding to the set of sampling points, wherein the contour generation model is obtained by training an initial back propagation network using an adaptive crossover and mutation differential evolution model; the problem of being unable to obtain a closed contour can be solved; and a closed contour can be obtained by an improved approximation polygon search model and a back propagation network fusion method.

[0005] For example, the patent application with the publication number CN110442994A discloses a 3D printing slicing processing method based on graph theory, which includes: 1) sorting the triangular mesh model data to reduce the number of intersections between the cutting plane and the triangle piece and improving the layer processing speed; 2) calculating the total number of slices of the model and the height of the cutting plane of each layer along the z-axis according to the slicing accuracy, and then screening the set of triangle pieces that intersect with the cutting plane of each layer; 3) calculating the intersection of the cutting plane and the set of triangle pieces, and constructing an undirected connected graph based on the connection relationship of the intersection; 4) calculating the connected components of the undirected connected graph constructed for each layer of slices, and sorting the vertices in each connected component based on the depth-first search algorithm, and then constructing the print filling area based on the sorting results and the closedness of the model. This invention uses a powerful search technology based on graph theory, without the need for complex calculations of mesh topological relationships and adjacency relationships, and can quickly obtain ordered layer cutting contours, which is suitable for 3D printing of closed and non-closed models.

[0006] The above patents all have the problem raised by this background technology: lack of refined analysis of the surface features of the model, making it difficult to accurately capture the key features of the model.

[0007] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgement or any form of suggestion that the information constitutes the prior art already known to ordinary technicians in this field. Summary of the invention

[0008] The technical problem to be solved by the present invention is to overcome the defects of the prior art and provide an intelligent slicing optimization method and system for a 3D printer, which automatically identifies model features and adjusts slicing parameters accordingly, effectively improves printing accuracy, reduces printing time, and reduces printing failure rate.

[0009] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0010] In one aspect, the present invention provides a method for optimizing intelligent slicing of a 3D printer, comprising the following steps:

[0011] S1: Obtain a 3D model to be sliced, and select feature points on the surface of the 3D model;

[0012] S2: extracting a feature index of each feature point, and calculating a precision value of each feature point based on the feature index;

[0013] S3: Segment the 3D model into sub-models based on the accuracy value of each feature point;

[0014] S4: Set a slicing plan for each sub-model based on the accuracy value of each feature point;

[0015] S5: Simulate slicing of each sub-model based on the slicing plan, and adjust the slicing plan based on the simulated slicing to generate a slicing scheme for each sub-model.

[0016] As a preferred solution of the intelligent slicing optimization method of the 3D printer of the present invention, wherein: the feature index includes curvature, roughness, and proximity; wherein the roughness is the variance of the heights of the p vertices closest to the feature point, and p is a positive integer; the height of the vertex is the distance of the vertex from the reference plane; the reference plane is a local plane fitted by the p vertices, and the heights of the vertices on both sides of the reference plane have opposite positive and negative relationships; the proximity is the average of the q smallest proximity distances of the feature point, and q is a positive integer; the proximity distance is the vertical distance of the feature point from any triangular mesh;

[0017] The calculation formula of the accuracy value of the feature point is as follows:

[0018] ;

[0019] Among them, f represents the precision value of any feature point; c represents the curvature of the feature point; s represents the roughness of the feature point; Indicates the proximity of feature points; Represents the weight coefficient.

[0020] As a preferred solution of the intelligent slicing optimization method of the 3D printer of the present invention, wherein: the method of selecting feature points on the surface of the 3D model is as follows: mapping the surface of the 3D model into a two-dimensional plane;

[0021] Selecting feature points on the two-dimensional plane at a fixed sampling interval using a space filling curve, and recording the coordinates of each feature point on the two-dimensional plane;

[0022] Based on the coordinates of each feature point on the two-dimensional plane, each feature point is mapped back to the surface of the 3D model.

[0023] As a preferred solution of the intelligent slicing optimization method of the 3D printer of the present invention, the method of dividing the 3D model into sub-models is as follows:

[0024] S301: Setting reference thresholds of m precision values ​​and building a reference threshold queue;

[0025] S302: Selecting the maximum reference threshold in the reference threshold queue as the accuracy threshold, and removing the accuracy threshold from the reference threshold queue;

[0026] S303: Determine whether the remaining model contains more than m segmentation feature points; the segmentation feature points are feature points whose precision values ​​are higher than the precision threshold; if so, execute step S304; if not, execute step S305; the remaining model is the unsegmented part of the 3D model;

[0027] S304: exploring a segmentation plane through an optimization algorithm, and segmenting a sub-model from the remaining model based on the segmentation plane;

[0028] S305: Repeat steps S302-S304 until the reference threshold queue is empty, and the segmentation of the 3D model is completed.

[0029] As a preferred solution of the intelligent slicing optimization method of the 3D printer of the present invention, the method of exploring the segmentation plane by the optimization algorithm and segmenting the sub-model from the remaining model based on the segmentation plane is as follows:

[0030] S3041: Explore a segmentation plane that maximizes the objective function through an optimization algorithm;

[0031] S3042: Determine whether the maximum objective function explored is greater than the objective function threshold, if so, execute step S3044, if not, execute step S3043;

[0032] S3043: exploring two segmentation planes that maximize the objective function through an optimization algorithm;

[0033] S3044: Based on the explored segmentation plane, a sub-model is segmented from the remaining model.

[0034] As a preferred solution of the intelligent slicing optimization method of the 3D printer of the present invention, the formula of the objective function is as follows:

[0035] ;

[0036] Where B represents the objective function; represents the number of segmentation feature points in the sub-model segmented by the segmentation plane; Represents the total number of feature points in the sub-model segmented by the segmentation plane; Represents the number of segmentation feature points in the remaining model before segmentation; Represents the volume of the sub-model separated by the splitting plane; Represents the total volume of the 3D model; Indicates the number of edges of the sub-model split by the splitting plane; Represents the number of edges of the remaining model after segmentation; represents the number of edges of the remaining model before segmentation; , , , are all weight coefficients.

[0037] As a preferred solution of the intelligent slicing optimization method of the 3D printer of the present invention, the slicing plan includes the slicing thickness of each sub-model, and the setting method is as follows:

[0038] S401: Calculate the mean of the precision values ​​of the feature points included in each sub-model;

[0039] S402: Obtain the number of types of printing materials involved in the sub-model;

[0040] S403: Calculate the slice thickness of each sub-model using the following formula:

[0041] ;

[0042] Where h represents the slice thickness of any sub-model; Represents the mean of the accuracy values ​​of all feature points on the sub-model; Indicates the number of types of printing materials involved in the sub-model; represents the weight coefficient; is the minimum threshold for slice thickness.

[0043] As a preferred solution of the intelligent slicing optimization method of the 3D printer of the present invention, the method of simulating slicing of each sub-model based on the slicing plan, adjusting the slicing plan based on the simulated slicing, and generating a slicing plan for each sub-model is as follows:

[0044] S501: simulating slicing of the sub-model based on the slice thickness of the sub-model in the slicing plan, and calculating and recording the slice area for each slice;

[0045] S502: Calculate the slicing error rate starting from the completion of the second layer simulation slicing, the formula is as follows:

[0046] ;

[0047] in, represents the slice error rate of the i-th slice, represents the slice area of ​​the i-th slice, represents the slice area of ​​the i-1th slice;

[0048] If the slicing error rate of the i-th layer of slicing is greater than the preset error rate threshold, the simulated slicing of the i-th layer is performed again, and the slice thickness of the i-th layer is reduced until the slicing error rate of the i-th layer of slicing is less than the preset error rate threshold, and the simulated slicing of the i+1-th layer is performed;

[0049] S503: Repeat step S502 until the sub-model is sliced; save the slice thickness of each layer of the sub-model slice as the slicing scheme.

[0050] In a second aspect, the present invention provides an intelligent slicing optimization system for a 3D printer, including a feature selection module, a feature extraction module, a segmentation module, a slicing plan module, and a slicing verification module; wherein:

[0051] The feature selection module is used to select feature points on the 3D model;

[0052] The feature extraction module is used to extract the feature index of each feature point and calculate the accuracy value of each feature point based on the feature index;

[0053] The segmentation module segments the 3D model into sub-models based on the accuracy value of each feature point;

[0054] The slicing plan module sets a slicing plan for each sub-model based on the accuracy value of each feature point;

[0055] The slicing verification module simulates slicing of each sub-model based on the slicing plan, and adjusts the slicing plan based on the simulated slicing to generate a slicing scheme for each sub-model.

[0056] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0057] By selecting feature points on the surface of the 3D model and calculating the precision value of each feature point, the present invention can automatically adjust the slice thickness of each sub-model, and realize the function of automatically adjusting the layer thickness according to the complexity and detail requirements of different areas. This means that thinner layer thickness is used in areas with rich details to improve precision, while thicker layer thickness is used in flat or simple areas to speed up printing, taking into account both printing precision and efficiency.

[0058] By exploring the optimal segmentation plane through the optimization algorithm, the present invention can intelligently segment the 3D model into multiple sub-models, ensuring that the accuracy values ​​of the feature points in each sub-model are relatively close, which is convenient for planning the slice thickness. By simulating the slice and monitoring the error rate of each slice layer, the step effect in the 3D printing process can be effectively reduced and the printing accuracy can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0060] Figure 1A flow chart of the intelligent slicing optimization method for a 3D printer provided by the present invention;

[0061] Figure 2 A schematic diagram of the structure of the intelligent slicing optimization system for a 3D printer provided by the present invention;

[0062] Figure 3 A flow chart of a method for segmenting a 3D model into sub-models provided by the present invention;

[0063] Figure 4 A flow chart of the method for exploring segmentation planes and segmenting sub-models provided by the present invention. DETAILED DESCRIPTION

[0064] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.

[0065] Example 1

[0066] This embodiment introduces a smart slicing optimization method for a 3D printer. Figure 1 , the method comprises the following steps:

[0067] S1: Obtain a 3D model to be sliced, and select feature points on the surface of the 3D model;

[0068] The way to select feature points is as follows:

[0069] Map the surface of a 3D model to a two-dimensional plane;

[0070] Selecting feature points on the two-dimensional plane at a fixed sampling interval using a space filling curve, and recording the coordinates of each feature point on the two-dimensional plane;

[0071] Based on the coordinates of each feature point on the two-dimensional plane, each feature point is mapped back to the surface of the 3D model.

[0072] Common space filling curves include Hilbert curves and Z-order curves, which can ensure uniform distribution of feature points. By uniformly selecting feature points on the two-dimensional plane of the 3D model, on the one hand, the coverage of feature points can be comprehensive; on the other hand, when the feature points are mapped back to the surface of the 3D model, the feature points in smoother areas are sparser, and the feature points in areas with more details are denser, ensuring that the selected feature points can fully reflect the characteristics of the 3D model surface.

[0073] S2: extracting a feature index of each feature point, and calculating a precision value of each feature point based on the feature index;

[0074] The characteristic indicators include curvature, roughness, and proximity; wherein the curvature indicates the degree of curvature of the model surface at the feature point. Most CAD software such as SolidWorks, Blender, etc. provide tools for calculating curvature. High curvature means that the model surface changes rapidly near this point, while low curvature means that the surface is relatively flat. When slicing, high curvature areas require thinner slice thicknesses to maintain details. The roughness is the variance of the heights of the p vertices closest to the feature point, where p is a positive integer; the height of the vertex is the distance of the vertex from the reference plane; the reference plane is a local plane fitted by p vertices, and the heights of the vertices on both sides of the reference plane have opposite positive and negative relationships; the reference plane can be fitted by methods such as the least squares method; the roughness describes the unevenness of the model surface. For areas with high surface roughness, a lower slice thickness is required to ensure the printing quality of the surface. The proximity is the average of the q smallest proximity distances of the feature points, where q is a positive integer; the proximity distance is the vertical distance between the feature point and any triangular mesh; if the proximity of a feature point is small, the slice thickness at that point should be set lower to avoid collision or fusion.

[0075] Many CAD software and 3D modeling tools represent and manipulate 3D models based on triangular meshes. For example, in Blender, you can use the Triangulate modifier to convert a model into a triangular mesh. A triangular mesh is a collection of connected triangles that approximate the surface of a 3D model. Each triangle consists of three vertices that define the position and shape of the triangle. Triangular meshes are the basis of many 3D modeling and rendering techniques.

[0076] The calculation formula of the accuracy value of the feature point is as follows:

[0077] ;

[0078] Among them, f represents the precision value of any feature point; c represents the curvature of the feature point; s represents the roughness of the feature point; Indicates the proximity of feature points; Represents a weight coefficient, which is set by technicians in this field based on actual needs.

[0079] S3: Based on the accuracy value of each feature point, the 3D model is divided into sub-models; the method is as follows Figure 3 , as follows:

[0080] S301: Setting reference thresholds of m precision values ​​and building a reference threshold queue;

[0081] S302: Selecting the maximum reference threshold in the reference threshold queue as the accuracy threshold, and removing the accuracy threshold from the reference threshold queue;

[0082] S303: Determine whether the remaining model contains more than m segmentation feature points; the segmentation feature points are feature points whose precision values ​​are higher than the precision threshold; if so, execute step S304; if not, execute step S305; the remaining model is the unsegmented part of the 3D model;

[0083] S304: Explore the segmentation plane through the optimization algorithm, and segment the sub-model from the remaining model based on the segmentation plane; the method is referenced Figure 4 , as follows:

[0084] S3041: A segmentation plane that maximizes the objective function is explored through an optimization algorithm. The formula of the objective function is as follows:

[0085] ;

[0086] Where B represents the objective function; represents the number of segmentation feature points in the sub-model segmented by the segmentation plane; Represents the total number of feature points in the sub-model segmented by the segmentation plane; Represents the number of segmentation feature points in the remaining model before segmentation; Represents the volume of the sub-model separated by the splitting plane; Represents the total volume of the 3D model; Indicates the number of edges of the sub-model split by the splitting plane; Represents the number of edges of the remaining model after segmentation; represents the number of edges of the remaining model before segmentation; , , , They are all weight coefficients, which are set by technicians in this field based on actual needs;

[0087] The above objective function encourages the optimization algorithm to explore the following characteristics of the segmentation plane: segment as many segmentation feature points as possible; in the segmented sub-models, the segmentation feature points account for a large proportion, and other feature points account for a small proportion; the volume of the segmented sub-model is as large as possible, because it is unacceptable to segment the model into dozens or hundreds of sub-models, and more segmentation times will result in printing preparation time and assembly costs that are much greater than the printing process itself; the number of newly added edges is small; Indicates the number of additional edges generated by the splitting plane when splitting the remaining model. The intersection of the splitting plane and the model may generate contours, that is, the number of additional edges; if multiple contours are generated, the splitting will produce multiple parts or smaller sub-models, which will have an adverse effect on the final assembly process.

[0088] S3042: Determine whether the maximum objective function explored is greater than the objective function threshold, if so, execute step S3044, if not, execute step S3043;

[0089] S3043: exploring two segmentation planes that maximize the objective function through an optimization algorithm;

[0090] If only one segmentation plane is used to segment the remaining model, it may not be possible to make the objective function greater than the preset objective function threshold, that is, using only one segmentation plane may not achieve a good segmentation effect. At this time, two segmentation planes are introduced to segment the remaining model, and the combination of the two segmentation planes that maximize the objective function is also found through the optimization algorithm, which can achieve a more accurate model segmentation than using one segmentation plane; however, if there are too many segmentation planes, it will cause too many new model features and parts. Therefore, in one segmentation, no more segmentation planes are introduced.

[0091] S3044: based on the explored segmentation plane, a sub-model is segmented from the remaining model; the segmented sub-model contains a large number of feature points with precision values ​​higher than the precision threshold, and a new remaining model is obtained (the remaining model before segmentation removes the sub-model segmented this time, and the remaining part, i.e., the remaining model after segmentation, is the new remaining model), and the sub-model no longer participates in subsequent segmentation;

[0092] S305: Repeat steps S302-S304 until the reference threshold queue is empty, and the segmentation of the 3D model is completed.

[0093] Dividing the model into several sub-models, the accuracy values ​​of the feature points contained in each sub-model are relatively close, making it easier to plan the slice thickness; printing each sub-model separately and assembling them into complete parts, or forming them into one piece on a multi-degree-of-freedom device by reasonably planning the printing sequence, can improve the manufacturing capabilities of complex models.

[0094] The optimization algorithm is an algorithm suitable for constrained optimization problems, such as genetic algorithm, particle swarm optimization, simulated annealing algorithm, etc. The process of exploring the segmentation plane through the optimization algorithm is as follows: define the search range of the segmentation plane, such as the normal vector of the plane and the distance between the plane and the origin; define the constraints of the segmentation plane, such as the plane cannot pass through certain specific parts of the 3D model; initialize the parameters required by the optimization algorithm, such as population size, number of iterations, etc.; implement the optimization algorithm, perform iterative optimization based on the objective function, and iterate until a predetermined termination condition is reached, such as the maximum number of iterations or the objective function is greater than a preset iteration threshold, etc.

[0095] S4: Set a slicing plan for each sub-model based on the accuracy value of each feature point;

[0096] The slicing plan includes the slice thickness of each sub-model, and the setting method is as follows:

[0097] S401: Calculate the mean of the precision values ​​of the feature points included in each sub-model;

[0098] S402: Obtain the number of types of printing materials involved in the sub-model;

[0099] S403: Calculate the slice thickness of each sub-model using the following formula:

[0100] ;

[0101] Where h represents the slice thickness of any sub-model; Represents the mean of the accuracy values ​​of all feature points on the sub-model; Indicates the number of types of printing materials involved in the sub-model; represents the weight coefficient, which is set by those skilled in the art based on experience; is the minimum threshold of slice thickness, which is set by technicians in this field based on experience and is related to the weight coefficient They work together to adjust the slice thickness h and ensure that its value is reasonable.

[0102] Based on the above slice thickness formula, the higher the accuracy value of the feature points of the sub-model, the smaller the slice thickness; the more types of printing materials involved, the smaller the slice thickness; the slice thickness can be automatically adjusted according to the complexity and detail requirements of the model, using thinner layer thickness in areas with rich details to improve accuracy, and using thicker layer thickness in flat or simple areas to speed up printing, taking into account both printing accuracy and efficiency.

[0103] In order to better illustrate the application of feature indicators, this embodiment is illustrated with a specific jewelry model. A complex jewelry pendant model includes fine patterns, inlaid beads, and some smooth surface areas. One side of the model is covered with complex patterns, which contain many tiny and delicate lines and other shapes. On the other side of the model, there are some raised beads with a certain gap between these beads. The back of the model is a relatively smooth plane without too many details. For the fine pattern part, due to its drastic surface changes, the curvature of the feature points is high, and a thinner slice thickness is required to maintain the details; for the smooth surface part, the curvature is low, and a thicker slice thickness can be used. At the same time, the roughness of the surface feature points in the fine pattern area is high, requiring a thinner slice thickness; the surface roughness of the smooth surface area is low, and a thicker slice thickness can be used. For the fine pattern and the inlaid beads, the proximity of the feature points is high, and a thinner slice thickness is required to avoid collision or fusion; for the smooth surface part, the proximity is low, and a thicker slice thickness can be used. Based on the above slice thickness calculation formula, the requirements for slice thickness setting in different areas of the jewelry model can be realized.

[0104] S5: simulate slicing each sub-model based on the slicing plan, and adjust the slicing plan based on the simulated slicing to generate a slicing scheme for each sub-model. The method is as follows:

[0105] S501: simulating slicing of the sub-model based on the slice thickness of the sub-model in the slicing plan, and calculating and recording the slice area for each slice;

[0106] S502: Calculate the slicing error rate starting from the completion of the second layer simulation slicing, the formula is as follows:

[0107] ;

[0108] in, represents the slice error rate of the i-th slice, represents the slice area of ​​the i-th slice, represents the slice area of ​​the i-1th slice;

[0109] If the slicing error rate of the i-th layer of slicing is greater than the preset error rate threshold, the simulated slicing of the i-th layer is performed again, and the slice thickness of the i-th layer is reduced until the slicing error rate of the i-th layer of slicing is less than the preset error rate threshold, and the simulated slicing of the i+1-th layer is performed;

[0110] S503: Repeat step S502 until the sub-model is sliced; save the slice thickness of each slice layer of the sub-model as the slicing scheme. By controlling the error rate of any slice layer to be no greater than the error rate threshold, the step effect of 3D printing can be effectively reduced and the printing accuracy can be improved.

[0111] Example 2

[0112] This embodiment is the second embodiment of the present invention; it is based on the same inventive concept as embodiment 1, Figure 2 This embodiment introduces an intelligent slicing optimization system for a 3D printer, including a feature selection module, a feature extraction module, a segmentation module, a slicing plan module, and a slicing verification module; wherein:

[0113] The feature selection module is used to select feature points on the 3D model; the module is equipped with a UV mapping program and a space filling curve, which can uniformly select feature points in two-dimensional space and map them back to the 3D model;

[0114] The feature extraction module is used to extract the feature index of each feature point; the feature index includes the curvature, roughness and proximity of each feature point; the module is also configured with a precision value calculation formula to calculate the precision value of each feature point based on the feature index;

[0115] The segmentation module segments the 3D model into sub-models based on the precision value of each feature point. The module is configured with a reference threshold of the precision value and an optimization algorithm, and can segment the 3D model into several sub-models, where the precision values ​​of the feature points contained in any sub-model are relatively close.

[0116] The slicing plan module sets a slicing plan for each sub-model based on the precision value of each feature point; the slicing plan includes the slice thickness of each sub-model; the module is configured with a slice thickness calculation formula, and can adaptively set the slice thickness for each sub-model based on the precision value of the feature point of the sub-model and the number of types of printing materials involved in the sub-model;

[0117] The slice verification module simulates slicing of each sub-model based on the slicing plan, adjusts the slicing plan based on the simulated slicing, and generates a slicing scheme for each sub-model; this module iteratively simulates slicing of each layer of the sub-model, calculates the error rate threshold of the slice to adjust the slice thickness, and generates the final slicing scheme.

[0118] The method of the present invention is not only applicable to general 3D printing applications, but also particularly applicable to fields that require high detail and precision, such as jewelry manufacturing. For applications such as jewelry manufacturing that require extremely high precision, the present invention can significantly improve the aesthetics of the printed parts.

[0119] The specific functions of the above modules are realized by referring to the relevant contents of the intelligent slicing optimization method for a 3D printer described in Example 1, which will not be elaborated here.

[0120] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0121] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the purpose and scope of protection of the present invention, which are all within the protection of the present invention.

Claims

1. The intelligent slicing optimization method of 3D printer is characterized by: The following steps are involved: S1: Obtain a 3D model to be sliced, and select feature points on the surface of the 3D model; S2: extracting a feature index of each feature point, and calculating a precision value of each feature point based on the feature index; The characteristic indexes include curvature, roughness, and proximity; wherein the roughness is the variance of the heights of the p vertices closest to the feature point, where p is a positive integer; the height of the vertex is the distance of the vertex from the reference plane; the reference plane is a local plane fitted by the p vertices, and the heights of the vertices on both sides of the reference plane have opposite positive and negative relationships; the proximity is the average of the q smallest proximity distances of the feature point, where q is a positive integer; the proximity distance is the vertical distance between the feature point and any triangular mesh; The calculation formula of the accuracy value of the feature point is as follows: ； Where f represents the precision value of any feature point; c represents the curvature of the feature point; s represents the roughness of the feature point; Indicates the proximity of feature points; represents the weight coefficient; S3: Segment the 3D model into sub-models based on the accuracy value of each feature point; The method for dividing the 3D model into sub-models is as follows: S301: Setting reference thresholds of m precision values ​​and building a reference threshold queue; S302: Selecting the maximum reference threshold in the reference threshold queue as the accuracy threshold, and removing the accuracy threshold from the reference threshold queue; S303: Determine whether the remaining model contains more than m segmentation feature points; the segmentation feature points are feature points whose precision values ​​are higher than the precision threshold; if so, execute step S304; if not, execute step S305; the remaining model is the unsegmented part of the 3D model; S304: exploring a segmentation plane through an optimization algorithm, and segmenting a sub-model from the remaining model based on the segmentation plane; S305: repeating steps S302-S304 until the reference threshold queue is empty, and the segmentation of the 3D model is completed; The method of exploring the segmentation plane through the optimization algorithm and segmenting the sub-model from the remaining model based on the segmentation plane is as follows: S3041: Explore a segmentation plane that maximizes the objective function through an optimization algorithm; S3042: Determine whether the maximum objective function explored is greater than the objective function threshold, if so, execute step S3044, if not, execute step S3043; S3043: exploring two segmentation planes that maximize the objective function through an optimization algorithm; S3044: Segmenting a sub-model from the remaining model based on the explored segmentation plane; The objective function formula is as follows: ； Where B represents the objective function; represents the number of segmentation feature points in the sub-model segmented by the segmentation plane; Represents the total number of feature points in the sub-model segmented by the segmentation plane; Represents the number of segmentation feature points in the remaining model before segmentation; Represents the volume of the sub-model separated by the splitting plane; Represents the total volume of the 3D model; Indicates the number of edges of the sub-model split by the splitting plane; Represents the number of edges of the remaining model after segmentation; represents the number of edges of the remaining model before segmentation; , , , All are weight coefficients; S4: Set a slicing plan for each sub-model based on the accuracy value of each feature point; S5: Simulate slicing of each sub-model based on the slicing plan, and adjust the slicing plan based on the simulated slicing to generate a slicing scheme for each sub-model.

2. The intelligent slicing optimization method for a 3D printer according to claim 1, characterized in that: The method for selecting feature points on the surface of the 3D model is as follows: mapping the surface of the 3D model into a two-dimensional plane; Selecting feature points on the two-dimensional plane at a fixed sampling interval using a space filling curve, and recording the coordinates of each feature point on the two-dimensional plane; Based on the coordinates of each feature point on the two-dimensional plane, each feature point is mapped back to the surface of the 3D model.

3. The intelligent slicing optimization method for a 3D printer according to claim 2, characterized in that: The slicing plan includes the slice thickness of each sub-model, and the setting method is as follows: S401: Calculate the mean of the precision values ​​of the feature points included in each sub-model; S402: Obtain the number of types of printing materials involved in the sub-model; S403: Calculate the slice thickness of each sub-model using the following formula: ； Where h represents the slice thickness of any sub-model; Represents the mean of the accuracy values ​​of all feature points on the sub-model; Indicates the number of types of printing materials involved in the sub-model; represents the weight coefficient; is the minimum threshold for slice thickness.

4. The intelligent slicing optimization method for a 3D printer as claimed in claim 3, characterized in that: The method of simulating slicing of each sub-model based on the slicing plan, adjusting the slicing plan based on the simulated slicing, and generating a slicing scheme for each sub-model is as follows: S501: simulating slicing of the sub-model based on the slice thickness of the sub-model in the slicing plan, and calculating and recording the slice area for each slice; S502: Calculate the slicing error rate starting from the completion of the second layer simulation slicing, the formula is as follows: ； in, represents the slice error rate of the i-th layer slice, represents the slice area of ​​the i-th slice, represents the slice area of ​​the i-1th slice; If the slicing error rate of the i-th layer of slicing is greater than the preset error rate threshold, the simulated slicing of the i-th layer is performed again, and the slice thickness of the i-th layer is reduced until the slicing error rate of the i-th layer of slicing is less than the preset error rate threshold, and the simulated slicing of the i+1-th layer is performed; S503: Repeat step S502 until the sub-model is sliced; save the slice thickness of each layer of the sub-model slice as the slicing scheme.

5. A 3D printer intelligent slicing optimization system, used to implement the intelligent slicing optimization method of a 3D printer according to any one of claims 1 to 4, characterized in that it comprises a feature selection module, a feature extraction module, a segmentation module, a slicing plan module, and a slicing verification module; wherein: The feature selection module is used to select feature points on the 3D model; The feature extraction module is used to extract the feature index of each feature point and calculate the accuracy value of each feature point based on the feature index; The segmentation module segments the 3D model into sub-models based on the accuracy value of each feature point; The slicing plan module sets a slicing plan for each sub-model based on the accuracy value of each feature point; The slicing verification module simulates slicing of each sub-model based on the slicing plan, and adjusts the slicing plan based on the simulated slicing to generate a slicing scheme for each sub-model.

Images