• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    Method and system for obtaining virtual geometries and detection of non-moldable areas in pieces, based on the meshing of a geometry (q) of a piece (200) in a plurality of points (pij) and facets (fi) triangular; and in the classification of said points (pij) and facets (fi) following an analysis by planes according to a direction of demolding (vector vz). For this, the facets (fi) are grouped into subgroups (πk) with which closed polygons are defined (ck). Each point (pij) is then classified according to its projection on the closed polygons (ck); and each facet (fi) is classified according to the classification of the points (pij) it contains. Thus, an automated and efficient analysis of any arbitrary morphology of the piece (200) is achieved. (Machine-translation by Google Translate, not legally binding)
    公开号:ES2594772A1
    申请号:ES201531899
    申请日:2015-12-23
    公开日:2016-12-22
    发明作者:Cristina MARTÍN DOÑATE；Miguel Ángel RUBIO PARAMIO；Antonio VIZÁN IDOIPE；Jorge Manuel MERCADO COLMENERO
    申请人:Universidad Politecnica de Madrid；Universidad de Jaen；
    IPC主号:
    专利说明:

    Method and system of obtaining virtual geometries and detection of non-moldable areas in parts
    Object of the invention
    The present invention relates to the field of manufacturing mechanical parts by means of molds, and more specifically to a system and method of obtaining virtual geometries and
    10 detection of non-moldable areas in parts.
    Background of the invention
    In all mold-based manufacturing techniques, such as injection of
    15 plastic materials or metal foundry, it is common to find a design phase in which it is necessary to quickly validate the workability of the pieces, detecting and differentiating small counter-outputs in an automated way, as well as areas that cannot be manufactured. The commercial systems currently used by companies for the resolution of the demoldability analysis involve a lot of manual interaction on the part
    20 of the designer, who is required not only knowledge of the particular tool, but also have experience in the manufacture of parts.
    Shorten the design and manufacturing times, establish a good precision and quality in the finishing of the piece, as well as be able to make changes in the design quickly are the
    25 main concerns of new product designers in the injection molding industry. The resolution of these problems requires a complete automation of the process of analysis of the plastic part in terms of its manufacture. Although there are many, related to the analysis of geometric desmoldeabilidad, these present quite a few disadvantages since they are either linked to the modeler analyzing the piece of
    30 internally, or need additional calculation devices such as graphic processing units (GPU), or are not valid for the analysis of all types of plastic parts.
    In particular, algorithms are known that use tessellated models as input to their
    35 procedure of demoulding analysis. The tessellated models of the plastic part provide a great advantage when the piece incorporates complicated surfaces and


    freely. Within this strategy, there are various approaches. For example, there are methods that identify non-convex regions by determining the set of possible demolding directions. The most appropriate demoulding direction among the possible ones is the one that provides the greatest value in a sum factor of several individual factors, among which are, visibility, flatness of the partition line and demoulding depth. According to a second example, a tessellated model can be used as a basis for classifying the surfaces of the casting, identifying counter-outputs and protrusions, finally generating a series of feasible partition lines for a given demolding direction. The most appropriate partition line is obtained by applying common use criteria in the company.
    Procedures based on tessellated models for generating lateral slide surfaces and smoothed partition lines have also been described. Given a series of counter-faceted facets in a polyhedral piece and the main demoulding direction, a set of movement address spaces can be computed for each facet in counter-output. This space represents the set of candidate translation vectors that can be used on the side slides. Likewise, a smoothed partition line can be computed that is placed through a band of triangles whose normal is approximately perpendicular to the demolding direction.
    The methods presented use a faceted model as a basis for the application of their counter-recognition procedures, however many of them assume that the piece to be manufactured is moldable, without internal slides or it should not have characteristics of counter-output (English ' undercut feature ') that interact. On the other hand these methods work with the normal to facets which makes the set of calculations complex and requires additional hardware.
    Other alternatives carry out the analysis of the demouldability of the piece through the study of the sections made to the solid non-tessellated model using a cutting procedure. This group includes a set of techniques based on making cuts to the solid model, which result in a variable set of sections, which are then crossed by a beam of straight lines. From the analysis of the points obtained, the study of demouldability is deduced. For example, a first approach analyzes the sections made to the plastic piece by establishing a set of rules based on experience and a vast bibliographic collection. This method can be improved by extending the scope to parts with partition lines


    with several contours and increasing the accuracy of the procedure. More recently, procedures based on discretization of the solid have been proposed, which transforms the solid body into a mesh of points, reallocating the points in the mesh based on their demouldability. Although the main advantage of these methods is that they analyze the geometry of the piece externally, they fail to achieve the accuracy of the methods based on the recognition of characteristics (of the English 'feature').
    According to some particular examples, US 8,296,097 B2 presents a technique based on elements rather than on features, including the calculation of a normal line to each surface of the piece and the analysis of the set of said lines. For its part, WO 2001027881 A2 shows an analysis by triangular meshing and calculation of displacement vectors that allow its manufacturing to be planned. Likewise, ES 2,512,940 A2 presents an automated validation method of the fabricability of three-dimensional object designs based on its geometry, which comprises transforming a geometric definition of a three-dimensional object that is desired to be manufactured in a three-dimensional cubic mesh and analyzing the object's fabricability three-dimensional based on the initial three-dimensional cubic mesh.
    However, all the methods described have limitations in terms of accuracy, computational requirements, a variety of analysable problems or automation. In short, there is still a need in the state of the art for a method and system of demolding analysis capable of adapting to any arbitrary morphology of the parts and providing an accurate and efficient demolding analysis, and minimizing the required manual interaction to user.
    Description of the invention
    The present invention solves the problems described above by means of a hybrid demoulding analysis technique, that is to say, combining node analysis and facet analysis, based on triangular meshing and analysis by planes perpendicular to the principal direction of demolding.
    In a first aspect of the invention a method of demolding analysis is presented which takes as input data a geometry of a piece to be manufactured by mold-based techniques, such as injection of plastic materials or metal smelting, as well as a direction main release of said piece. The address of


    demoulding can be introduced by a user or obtained from other sources, such as a computer-readable medium, an exit from an automated optimization process, etc. The starting geometry can be recovered from the aforementioned computer-readable medium or from a separate memory.
    The model takes as a starting point the meshed geometry consisting of a plurality of triangular nodes and facets. Next, the facets are grouped into a plurality of subgroups, so that each subgroup comprises those facets with at least one of its three nodes in the same analysis plane, and said analysis planes being defined as planes perpendicular to the direction of demoulding Note that the same facet can be grouped into several different subgroups. Note also that particular embodiments of the method of the invention may comprise a previous step of generating the mesh geometry described.
    Once the facets are grouped, a classification of the nodes in demolding is carried out in a positive direction according to the demoulding direction (which we will simply call "positive sense" for simplicity), unmolding in the negative direction according to the demolding direction (which we will call "negative direction "for simplicity) and non-releaseable through a plane analysis. Said plane analysis comprises defining a closed polygonal contour from the facets of each subgroup associated with an analysis plane, and the projection of each node on said closed polygonal contours. Preferably, the analysis is carried out in the demoulding direction in the positive direction and in the negative direction, projecting each point on the immediately preceding plane in said direction. Also, polygonal contours are defined as the projection of a subset of facets on the plane to which they are associated. More preferably, in the case of the analysis following the positive direction of the demoulding direction, if the projection of a node is strictly outside the closed polygon, said node is classified as demoldable in the demoulding direction, being further moldable by the upper cavity of the mold. If, on the contrary, its projection is inside or the border of the closed polygon, it is classified as non-removable. This process is repeated in the direction of demolding in the negative direction, thus distinguishing the demolding nodes in the direction of demolding in the negative direction within those previously classified as non-demolding.
    Once the nodes have been classified, this classification serves as a basis for classifying the facets that comprise them. Said facets are preferably classified in the following 5


    categories:
     Removable in the direction of demolding in the positive direction (moldable by upper cavity), when all its nodes are demolding in said direction and direction
    5  Removable in the direction of demolding in the negative direction (moldable by lower cavity), when all its nodes are demolding in said direction and direction.
     Non-releaseable or partially releaseable, when at least one of its nodes is classified as non-releaseable. To distinguish between facets not
    10 releasable or partially demolding, preferably a facet-to-facet analysis based on the dimensions in the axis of the demoulding direction of the nodes that conform them is applied.
     Preferably, the nodes can also be classified as laterally releasable by repeating the analysis described for an analysis direction rotated 90 ° 15 with respect to the main demolding direction.
    Preferably, the method comprises calculating a partition line as a separation between demouldable facets by upper and lower cavity. Said calculation of the partition line may also comprise dividing the partially demoldable facets into
    20 a releaseable subregion and a non-releaseable subregion.
    Finally, the classification of facets carried out is stored in a computer-readable medium, and can also be shown to the user, who can choose a new direction of demolding based on them.
    In a second aspect of the invention there is presented a demoulding analysis system that implements the steps of the described method, which may include any characteristic of any of the preferred options or particular embodiments of said method. In particular, the system comprises:
    30  One or more computer-readable storage media, in which the starting tessellated geometry, the results of the demoulding analysis and, potentially, any other information used by said analysis are stored.
     In some particular embodiments, the method may also comprise a meshing module, which performs the division of the starting geometry into triangular facets.  A grouping module, which groups in subgroups all those facets that 6


    they comprise nodes in the same analysis plane perpendicular to the demoulding direction.  A node classification module that performs the classification into demoldable nodes by upper cavity, lower cavity and non-demouldable following the plane analysis described for the method of the invention.
     A facet classification module that performs the classification into facets that can be demoulded by the upper cavity, lower cavity, non-demolding and, preferably, can be demoulded laterally, based on the classification of the nodes they comprise, following the analysis described for the method of
    10 invention.  Preferably, a calculation module that calculates the partition line between upper and lower cavity, following the analysis described for the method of the invention.
     Control means that store the results obtained and coordinate the operation of the rest of the system modules in the storage medium readable by computer.
    Finally, in a third aspect of the invention a computer program is presented comprising computer program code means adapted to implement the
    The described method, when running on a computer, a digital signal processor, an application-specific integrated circuit, a microprocessor, a microcontroller or any other form of programmable hardware.
    The system, method, and computer program described therefore provide a
    25 automated, accurate and efficient demoulding analysis tool; applicable to any geometry and internal constitution of the piece under analysis. Likewise, it is possible to shorten the design and manufacturing times, establish good precision and quality in the finishing of the piece, as well as be able to make changes in the design quickly. Finally, the proposed technique is independent of the model of the piece, does not require a model
    30 analytical of the contour surface of the piece and does not require additional calculation devices. These and other advantages of the invention will be apparent in light of the detailed description thereof.
    Description of the figures
    In order to help a better understanding of the features of the invention of 7


    according to a preferred example of practical realization thereof, and to complement this description, the following figures are attached as an integral part thereof, the character of which is illustrative and not limiting:
    5 Figure 1 schematically shows the main components of an implementationparticular of the system of the invention and the information exchanged between them.
    Figure 2 shows the geometry of an example piece on which a particular implementation of the method of the invention is applied.
    Figure 3 illustrates the meshed model of the example piece.
    Figure 4 is an enlarged view of said meshing of the example piece.
    Figure 5 exemplifies the classification in planes made by a particular implementation of the method of the invention following the main direction of demolding.
    Figure 6 shows the grouping of flat to flat facets according to a particular implementation of the method of the invention.
    Figure 7 shows the classification of facets resulting from applying a particular implementation of the method of the invention to the example piece.
    Preferred Embodiment of the Invention
    25 In this text, the term "comprises" and its derivations (such as "understanding", etc.) should not be understood in an exclusive sense, that is, these terms should not be construed as excluding the possibility that what is described and define can include more elements, stages, etc.
    Note that, in order to assist in the understanding of the invention, reference is made in the present invention to matrices and vectors in which the relevant information used by it is stored. However, the person skilled in the art will understand that there may be alternative ways of structuring and storing such information within the scope of the
    Invention as claimed. Note also that, for simplicity, the notation of these matrices has been used to reference the information in text and figures 8


    contained within these matrices.
    Figure 1 shows the main modules of a preferred embodiment of the system of the invention, which in turn implement a preferred embodiment of the method of the invention. Likewise, all the steps of the method of the invention can be carried out by means of computer code processing, all the calculations involved in meshing and the classification of nodes and facets being able to be performed in a computer processor, a digital signal processor, an application-specific integrated circuit, a microprocessor, a microcontroller or any other form of programmable hardware. Note that although each module has been represented independently, note that there may be particular implementations in which several modules are integrated into the same element, as well as embodiments in which one module is subdivided into several. Likewise, the main information exchanged by these modules has been shown. However, such representation does not exclude that the modules exchange any other additional information. In the same way, the exchange of information between modules can be done through common elements, such as memories shared by several modules, instead of being transmitted directly.
    The system comprises an input / output interface (101) that receives information from the user and shows the results obtained. The interface may be implemented in accordance with any technique or combination of techniques known in the state of the art, such as screens, keyboards, touch screens, etc. As the main information, the user enters the main demoulding address (V⃗z) that serves as the input parameter of the system, and receives the geometry information with integrated demoulding analysis (Q '' ') either in its version with node information (Q '' 'Nn), in its version with facet information (Q' '' Ff) or both. Note that the interface (101) can receive any other command or additional information from the user relevant to the demoulding analysis process. Also, note that in particular implementations of the invention, the demoulding direction (V⃗z) can be stored together with the geometry (Q) of the part (200), or calculated automatically by the system itself
    The interface (101) transmits the information entered by the user to the control means (102), which coordinate the operation of the rest of the system elements and store the results obtained by the same in a computer readable storage medium (103). , such as geometry information with integrated demoulding analysis (Q '' '). Said computer readable storage medium (103) may comprise


    also the geometry information (Q) that serves as the starting node of the procedure, or both functions can be divided into independent memories.
    The geometry (Q), extracted from the computer readable storage medium (103), is
    5 first received by the mesh module (104) that provides segmentationin nodes (Pij) and triangular facets (Fi), represented by the meshed node geometry(Q'n) and meshed facet geometry (Q'f). From these meshed geometries, and indemoulding direction function (V⃗z) provided by the control means (102), agrouping module (105) determines a plurality of analysis planes (Plk)
    10 perpendicular to the demoulding direction (V⃗z) and a plurality of subgroups (Πk) of facets (Fi) associated with said planes, following the criteria detailed later in the present description.
    Next, a node classification module (106) first classifies nodes 15 (Pij) into the following categories:  Demouldable in the direction of demolding in the positive direction (Ωn1), which we will call the upper cavity for simplicity.  Removable in the direction of demolding in the negative direction (ξn3), which we will call the lower cavity for simplicity. 20  Non-releaseable (βn2).
    Depending on the classification of nodes (Pij) performed, a facet classification module
    (107) classify the facets into the following categories:  Demouldable in the direction of demolding in the positive direction (Ωf1), which we will call the upper cavity for simplicity.  Releaseable in the direction of demolding in the negative direction (ξf2), which
    We will call lower cavity for simplicity. Partially removable (Γf3). Non-releaseable (Δf4).
    30  Laterally removable (τf5), also called lateral slides in this document.
    The facet classification is finally used by the calculation module (108) to determine the partition line between the upper cavity and the lower cavity, as well as the
    35 non-removable areas. This information overlaps the original geometry (Q) giving


    place to geometry with integrated demoulding analysis (Q ’'', virtual geometry), which is stored in the computer readable storage medium (103). This storage can for example be carried out directly, or managed from the control means (102).
    Figure 2 shows the geometry (Q) of an example piece (200) used to describe the method of the invention. Note, however, that the method is valid for any arbitrary geometry. Figure 3 shows the same piece (200) after the mesh is applied in triangular facets, obtaining a discrete geometry by the meshed geometry of nodes (Q'n) and the meshed geometry of facets (Q'f). The meshed node geometry (Q'n) can be stored in a range matrix (nx3) that groups the entire set of triangular facets (where n is the total number of facet of the mesh), while the meshed facet geometry (Q 'f) can be stored in a 3nx3 range matrix that groups the entire set of nodes (Pij) of the mesh. A definition of the problem is thus achieved under independent analysis of the modeling of any other software or related method.
    During a first preprocessing phase, the nodes (Pij) of the geometric mesh are classified according to their dimension (zij) according to the direction vector (V⃗z) parallel to the demoulding direction and in the positive direction. The main demoulding address is determined by the user as the starting data, so that once the analysis has been carried out, said user can change his choice if he deems it convenient. Alternatively, in other particular implementations of the invention, the demoulding direction can be optimized in an automated manner based on the output of the algorithm.
    The set of dimensions (zij) of the nodes belonging to the mesh are stored in a one-dimensional vector arranged in ascending direction in the direction of demolding. A set of analysis planes (Plk) parallel to the XOY plane and perpendicular to the demolding direction are assigned to each value of the one-dimensional vector, as shown in Figure 5.
    Next, the facets (Fi) of the mesh are classified into levels or subgroups, within a multidimensional matrix (Πk) where each dimension k of it corresponds to an analysis plane (Plk), so that a dimension k It includes all those facets that meet the condition that at least one of its vertices belongs to the analysis plane. An example of this classification is shown in Figure 6.


    After grouping the different facets (Fi) according to the dimension value (zij) of their respective nodes, along the demoulding direction, the mesh is processed for its manufacturing analysis by geometrically analyzing the facets (Fi ) and the nodes (Pij) according to criteria of demoulding. To do this, the discrete geometry (Q ’) is evaluated according to the direction of demolding in the positive direction (V⃗z), with the aim of estimating the geometric regions of the piece that are manufactured by the upper cavity. Subsequently, a reorientation of the piece in the opposite direction to that previously analyzed allows to locate the regions that belong to the lower cavity.
    Initially, the different facets (Fi) and nodes (Pij) contained in each of the levels k (submatrices) of the multidimensional matrix (Πk) are selected, so, for a sequential plane (Plk) a closed polygonal contour is defined (Ck) of analysis from the projection of the facets contained in the level corresponding to the sequential plane (Plk) of analysis of the multidimensional matrix. The procedure projects the set of nodes (Pij) of the mesh of the piece on the closed contour (Ck) immediately above, checking if all the nodes can unmold along the direction of demolding and in the given direction. In this way, if a node (Pij) is strictly outside the study region, it is cataloged as an upper demoulding element, stored in a matrix Ωn1 responsible for storing the set of demouldable nodes in the positive direction of the demolding direction, by the On the contrary, if an element remains in the interior region or at the border, said node is classified as non-demolding, being stored in a βn2 matrix responsible for storing the set of non-demolding elements in the demolding direction. Thus:
    Q’n = {Ωn1∪βn2}
    with n1 + n2 = n.
    Note that a Pij node is a border node of Ck if every Pij environment contains at least nodes comprised in Ck and nodes not comprised in Ck. The set of border nodes of Ck is called the border of Ck and it shows Fr (Ck).
    That is, for the levels k between the second and the last, a series of control nodes or Gauss nodes (PGauss) are generated in all facets comprised in the closed contour (Ck). The algorithm projects the set of nodes {(Pij ∈ B’n) - (Pij ∈ Plk)} → zij <zk and the
    corresponding control nodes (PGauss) on Ck-1; checking whether the nodes Pij and PGauss can unmold along the direction of demolding (V⃗z) and given direction. Of this 12


    If a node (Pij) is strictly outside the study region Ck-1, it is cataloged as a superior release element included in Ωn1.
    After the execution of the k plane analysis, the nodes stored in the matrices responsible for storing the set of demouldable nodes in the positive and negative direction of the demoulding direction are compared. It may happen that a node has been cataloged as demoldable belonging to the matrix of demolding nodes according to the positive direction of the demolding and non-releasing direction belonging to the matrix of demolding nodes in a negative direction in different planes. That is why the elements of the matrices are compared and those nodes that have duplicity of behavior are cataloged again as non-demolding.
    Once the demoldability of the nodes (Pij) of the mesh is analyzed, the Fi facets are cataloged in demolding, non-demolding and partially demolding or semi-demolding forms, storing them in matrices with geometric information on the manufacturing of the piece. Firstly, the demouldability of the facets is cataloged along the positive direction of demoulding, serving as a base of cataloging the nodes previously classified in the matrix of superiorly demouldable nodes (Ωn1). All facets (Fi) that meet the condition that each and every one of its vertices belong to the demoldable region (Ωn1), are cataloged as demolding and assigned to the matrix of direct demoulding facets (Ωf1), as well as removed from the matrix Q'f.
    The remaining facets (Fi) belonging to the matrix Q'f are cataloged as non-demoulding, or semi-removable, facets along the + d direction, being assigned to the matrix formed by the union of the matrices Γf3 and Δf4. so that:
    Q’f = {Ωf1 ∪ ξf2 ∪ Γf3 ∪ Δf4}
    with f1 + f2 + f3 + f4 = f. We will notice as Ωf1, ξf2 the geometric places corresponding to the set of demouldable facets by upper cavity and lower cavity respectively, along the demoulding direction, and f1 and f2 being the number of facets stored in the matrices Ωf1 and ξf2 respectively. Likewise, we will notice as Γf3 the geometric place of the set of partially removable facets along the demolding direction, where f3 is the number of facets stored in the matrix Γf3; and as Δf4 the geometric place of the set of non-demoulding facets along the demolding direction, where f4 is the number of facets stored in the matrix Δf4.


    As a result of the application of the facet cataloging procedure in the mesh described up to this node, the horizontal and inclined Fi facets, removable according to the upper cavity, have been classified and stored in the Ωf1 matrix. However, there is still a set of facets classified as belonging to [Γf3 ∪ Δf4], which, due to their geometric configuration, can be reclassified as demoulding facets. These facets meet the geometric condition of perpendicularity with the direction of
    demoulding The Fi3.4 ∈ [Γf3 ∪Δf4] facets located under the border of a Fi1 ∈ Ωf1 facet, such
    that the projection of the nodes {Pi3.4 1, Pi3.4 2, Pi3.4 3} that make up the Fi3.4 facet, belongs to the border Fr (Fi1), are removable, being repositioned in the matrix Ωf1 and eliminated of the matrix [Γf3 ∪Δf4] where they were stored.
    In the case that some facets Fi ∈ Π1 form a coplanar region, they are included in Ωf1
    those facets Fi ∈ Pl1 coplanar to each other. The rest of Fi ∈ Q’f facets that have not been
    cataloged after the application of the procedure of reallocation of vertical facets as demoulding facets, they are stored in the matrix of non-demoulding and semi-removable facets [Γf3 ∪Δf4].
    Then, the same procedures set forth are applied again, but reorienting the part in the negative direction of the demoulding direction. This results in the set of demoulding facets corresponding to the region of the lower cavity (ξf2).
    To complete the classification, a series of unification criteria are established for those facets with double results:
     All those facets that belong to the releasable region corresponding to the upper and lower cavity will only be stored in the matrix ξf2 (lower cavity), being eliminated from the matrix Ωf1 (upper cavity).
     All those facets that belong to the releasable region corresponding to the lower cavity ξf2 and have been cataloged as non-removable ∈ [Γf3 ∪Δf4] in the
    First analysis (corresponding to + V⃗z), are stored in the matrix ξf2 (cavity


    lower), being removed from the matrix [Γf3 ∪Δf4].
     Similarly, all those facets that belong to the releasable region corresponding to the upper cavity Ωf1 and in turn have been cataloged as non-removable ∈ [Γf3 ∪Δf4] in the second analysis, will only be stored in
    the matrix Ωf1 (upper cavity), being eliminated from the matrix [Γf3 ∪Δf4].
    In this node of the mesh demoulding analysis, the facets Fi3.4 ∈ [Γf3 ∪Δf4] must
    be reclassified, in non-releaseable facets Fi4 ∈ Δf4 and semi-removable faces Fi3 ∈ Γf3.
    These facets meet the geometric condition of perpendicularity with the demoulding direction. For the resolution of this problem, the procedure uses an analysis methodology based on the overlap between the different non-demoulding and semi-removable facets.
    The application of the reclassification procedure results in a reclassification of the facets in matrices, in addition to dividing the semi-demountable facets into regions based on their demouldability. Following the application of the reclassification procedure, a new set of facets and nodes is created that make up a new virtual geometry with geometric and manufacturing information that complements the Q’f and Q’n meshes. For it,
    the procedure performs a facet-facet comparative analysis (Fis - Fir) ∈- [Γf3 ∪ Δf4] with the
    objective of classifying the overlapping facets and the facets that generate overlap.
    Given a pair of facets (Fis - Fir) ∈ [Γf3 ∪ Δf4], it is checked by comparing the
    vertical dimensions of its nodes, which of the facets produces overlap (it is in a higher plane) and if there is overlap. To check if there is an overlap between a pair of facets belonging to [Γf3 ∪ Δf4], both facets are projected on a plane
    perpendicular to the demoulding direction and it is checked by a logical Boolean operation if there is contact between these facets. That facet Fis ∈ [Γf3 ∪ Δf4] that
    it meets the condition that the z-dimension of its three nodes is greater than the z-dimension of the three nodes of the Fir facet and the intersection between facets is non-zero, is assigned to the matrix of non-demouldable facets Δf4. The other facet Fir is assigned to the matrix of semi-removable facets Γf3. This defines which facets belong to the set of


    semi-removable facets Γf3 and which belong to the set of non-demouldable facets Δf4, thus allowing to detect non-demoulding areas of the piece (200).
    The partition line (PL), by definition, is located in the boundary region between the upper cavity (Ωf1) and the lower cavity (ξf2). However, for those geometries (Q) for which there are semi-moldable facets between both cavities, a subalgorithm is proposed based on the decomposition of said semi-moldable facets. One time
    Once the Fi3 ∈ Γf3 facets have been defined with semi-removable behavior, they proceed to their
    Fragmentation finding for each facet the releasable region by upper or lower cavity and the non-moldable region. The division of the semi-removable facets is done by applying a subtraction and intersection methodology between each of the facets
    semi-removable ∈ Γf3 and the closed set of non-moldable facets (Δf4).
    In this way a Boolean subtraction operation is performed between the orthogonal projection of the facets perpendicular to the demoulding direction ∈ [Ωf1 ∪ ξf2] and the
    facets ∈ ξf2. As a result of this operation virtual polygonal divisions are generated
    F′′i1 ∈Ωf1, F′′i2 ∈ξf2 that allow to define the optimal partition line. That is to say:
    ∀Fi1∈ Ωf1; ∀Fi2∈ξf2: (Fi1, Fi2 ⊥ V⃗z → {Proj (Fi1) - ξf2 = F′′i1; Proj (Fi2) - ξf2 = F′′i2}
    F′′i1 ∈ Ωf1
    F′′i2 ∈ξf2
    Finally, the partition line is defined as.
    ∀Pij = {xij, yij, zij} ∶ (Pij ∈ Fr (Ωf1)) ∨ (Pij ∈ Fr (ξf2)) → Pij ∈ [PL]
    The closed set {Fi4 ∈ Δf4} and the semi-removable facets Fi3 ∈ Γf3 are projected onto
    a horizontal plane perpendicular to the demolding direction calculating the intersection and subtraction resulting from said elements. This geometric information is transferred back to the plane of the semi-removable facets under study. The result of the Boolean operation of intersection between the semi-removable Fi3 facets and the non-demouldable Fi4 facets causes for each facet its division into two regions, the


    releasable region F′′i1 ∈ Ωf1 and the non-removable region F′′i4 ∈Δf4.
    The set of facets {F''i4, F''i1} that after the execution of the reclassification procedure have been divided, depending on their demoldability, form a new virtual geometry Q''f that contains, the new set of regions polygonal virtual (F''i4, F''i1) that
    overlap on the Q’f mesh. The new mesh Q ’’ ’Ff = {Q’’f f Q’f} incorporates information
    on the release of the piece (200). The new virtual geometry therefore incorporates geometric information about manufacturing that is included in the resulting files and models. Similarly, fragmentation of facets creates new virtual nodes that are part of the virtual geometry Q’’n. Q’’n overlaps the Q’n mesh by adding information about the workability of the piece, forming the Q’ ’’ Nn mesh.
    After applying the face reclassification procedure in the positive direction of the demoulding direction, it is applied analogously in the reverse direction of the demoulding direction obtaining the semi-removable facets ∈ Γf3 and ∈ Δf4 in this direction.
    Finally, the set of facets classified as non-moldable F′′i4 ∈ Δf4 incorporates a
    group of facets F ’’ ’5 that can be solvable by lateral sliding in the direction perpendicular to the demolding direction. To properly classify these facets, the analysis is repeated again after reorienting the problem 90º with respect to the X axis and subsequently with respect to the Y axis, excluding the regions previously classified as
    releasable ∈ [Ωf1 ∪ ξf2]. That is, the analysis is repeated considering a direction of
    Auxiliary demolding perpendicular to the previous demoulding direction (V⃗z). The releasing Fi5 facets along the direction perpendicular to the demolding direction are stored in the matrix τf5. The final result of the procedure is indicated in the example figure 7.
    In view of this description and figures, the person skilled in the art may understand that the invention has been described according to some preferred embodiments thereof, but that multiple variations can be introduced in said preferred embodiments, without departing from the object of the invention such and as claimed.
    Note that all investigations on which the patent is based have been carried out and


    financed within the project entitled: ”Design of a vertical software for the integration of the operations of automated analysis of demoulding, tooling design and cost estimation in molded parts by plastic injection (CELERMOLD) financed by the Ministry of Economy, Science and Employment (Junta de Andalucía) (Project code TI-12 TIC-1623).

    权利要求:
    Claims (11)
    [1]
    1. Manufacturing method based on obtaining virtual geometries and detecting
    non-moldable areas in parts, given a geometry (Q) of a piece (200) and a main demoulding direction (V⃗z), characterized in that the method comprises:
    i. recovering from a computer-readable storage medium (103), the geometry (Q) of the piece divided into a plurality of triangular nodes (Pij) and facets (Fi) obtained by means of a scanner scanning process;
    ii. obtain the demoulding address (V⃗z) by entering it by a user;
    10 iii. group the facets (Fi) into a plurality of subgroups (Πk), those facets (Fi) that comprise at least one node (Pij) in the same plane (Plk) perpendicular to a demolding direction (V⃗z) belonging to the same subgroup. ;
    iv. classify the plurality of nodes (Pij) in demolding in the direction of demolding (Ωn1) in the positive direction, demolding in the direction of demolding
    15 in the negative direction (ξn3) and non-demoulding (βn2) depending on at least one projection of each node (Pij) on a closed polygonal contour (Ck) defined by the facets (Fi) of a subgroup (Πk);
    v. classify the plurality of facets (Fi) in mold release in the direction of demolding in the positive direction (Ωf1), release in the direction of demolding
    20 in the negative direction (ξf2) partially demolding (Γf3) and non-demolding (Δf4) based on, at least, the classification of the nodes (Pij) of each facet (Fi); Y
    saw. Store the facet classification (Fi) in the computer readable storage medium (103).
    vii. manufacture a real piece using a demoulding direction based on the facet classification performed.
    [2]
    2. Method according to claim 1 characterized in that the step of classifying the plurality of nodes (Pij) comprises:  calculating the closed polygonal contour (Ck) as the projection of a subgroup of
    30 (Πk) facets (Fi) on a plane (Plk); and  project each node (Pij) on the closed polygonal contour (Ck) of the closest anterior plane (Plk) in the demolding direction (V⃗z).
    [3]
    3. Method according to claim 2 characterized in that the step of classifying the plurality of nodes (Pij) further comprises:
     if the projection of the node (Pij) is strictly outside the closed polygon (Ck), 19

    classify the node (Pij) as demolding in the demoulding direction (Ωn1); Y
     if the projection of the node (Pij) is in an inner region or a boundary of the closed polygon (Ck), classify the node (Pij) as non-demolding (βn2).
    Method according to any one of the preceding claims characterized in that the step of classifying the plurality of facets (Fi) comprises:  if all the nodes of a facet (Fi) are classified as demolding in the
    demoulding direction (Ωn1), classify the facet (Fi) as releasable in the demoulding direction (Ωf1);
    10  if all the nodes of a facet (Fi) are classified as demolding in the direction of demolding in the negative direction (ξn3), classify the facet (Fi) as demolding in the direction opposite to the direction of demolding (ξf2); Y
     if a facet (Fi) comprises at least one node (Pij) classified as non-releaseable (βn2), classify the facet (Fi) as partially demoldable (Γf3) 15 or as non-releaseable (Δf4) based on a facet analysis a facet based on
    the dimensions of its nodes.
    [5]
    5. Method according to any of the preceding claims characterized in that it comprises:  analyzing the nodes (Pij) and facets (Fi) according to at least one auxiliary direction rotated 90 ° with respect to the demoulding direction (V⃗z); Y
     reclassify as laterally demouldable facets (τf5) the facets (Fi) previously classified as non-demolding (Δf4) and partially demolding (Γf3) that are demolding according to the auxiliary direction.
    [6]
    Method according to any of the preceding claims, characterized in that it comprises calculating a partition line as a separation between demoulding facets (Fi) in the demoulding direction (Ωf1) and demoulding facets (Fi) in the direction opposite to the demolding direction. (ξf2).
    [7]
    7. Method according to the claim characterized in that the step of calculating the partition line comprises subdividing each partially demouldable facet (Γf3) into a first releasable region and a second non-demouldable region.
    [8]
    8. System for obtaining virtual geometries and detection of non-moldable areas
    in pieces, given a geometry (Q) of a piece (200) and a demoulding direction 20

    (V⃗z) main, characterized in that the system comprises:
     a computer readable storage medium (103) that provides a geometry (Q) of a part divided into a plurality of triangular nodes (Pij) and facets (Fi);
    5  grouping module (105) configured to group the facets (Fi) into a plurality of subgroups (Πk), those facets (Fi) belonging to at least one node (Pij) in the same plane (Plk) belonging to the same subgroup ) perpendicular to a demolding direction (V⃗z) entered by the user;
     node classification module (106) configured to classify the plurality of
    10 nodes (Pij) in demolding in the direction of demolding in the positive direction (Ωn1), demolding in the direction of demolding in the negative direction (ξn3) and non-molding (βn2) depending on at least one projection of each node ( Pij) on a closed polygonal contour (Ck) defined by the facets (Fi) of a subgroup (Πk);
    15  facet classification module (107) configured to classify the plurality of facets (Fi) in demolding in the direction of demolding in the positive direction (Ωf1), demolding in the direction of demolding in the negative direction (ξf2) partially demolding (Γf3 ) and non-removable (Δf4) based on, at least, the classification of the nodes (Pij) of each facet (Fi); Y
    20  control means (102) configured to store the facet classification (Fi) in the computer readable storage medium (103).
    [9]
    9. System according to claim 8 characterized in that the node classification module (106) is further configured to: 25  calculate the closed polygonal contour (Ck) as the projection of a subset of (Πk) facets (Fi) on a flat (Plk); Y
     project each node (Pij) on the closed polygonal contour (Ck) of the closest previous plane (Plk) in the demolding direction (V⃗z).
    System according to claim 9 characterized in that the node classification module (106) is further configured to:  if the projection of the node (Pij) is strictly outside the closed polygon (Ck),
    classify the node (Pij) as demolding in the demoulding direction (Ωn1); and  if the projection of the node (Pij) is in an inner region or a border of the closed polygon (Ck), classify the node (Pij) as non-demolding (βn2).

    [11]
    11. System according to any of claims 8 to 10 characterized in that the facet classification module (107) is further configured to:  if all the nodes of a facet (Fi) are classified as demolding in the
    5 demoulding direction (Ωn1), classify the facet (Fi) as demolding in the demoulding direction (Ωf1);
     if all the nodes of a facet (Fi) are classified as demolding in the direction of demolding in the negative direction (ξn3), classify the facet (Fi) as demolding in the direction opposite to the direction of demolding (ξf2); Y
    10  if a facet (Fi) comprises at least one node (Pij) classified as non-releasable (βn2), classify the facet (Fi) as partially demoldable (Γf3)
    or as non-releaseable (Δf4) based on a facet-to-facet analysis based on the dimensions of its nodes.
    A system according to any of claims 8 to 11 characterized in that the node classification module (106) and the facet classification module (107) are further configured to:  analyze the nodes (Pij) and facets ( Fi) according to at least one auxiliary direction rotated 90 ° with respect to the demoulding direction (V⃗z); Y
    20  reclassify as laterally demoulding facets (τf5) the facets (Fi) previously classified as non-demolding (Δf4) and partially demolding (Γf3) that are demolding according to the auxiliary direction.
    [13]
    13. System according to any of claims 8 to 12 characterized in that
    25 comprises a calculation module (108) configured to calculate a partition line as a separation between demoulding facets (Fi) in the demoulding direction (Ωf1) and demoulding facets (Fi) in the direction opposite to the demolding direction (ξf2 ).
    A system according to claim 13, characterized in that the calculation module (109) is further configured to subdivide each partially demouldable facet (Γf3) into a first releasable region and a second non-demouldable region.
    [15]
    15. Computer program comprising program code means of 35 computer adapted to perform the steps of the method according to any
    of claims 1 to 7, when said program is executed in a

    computer, a digital signal processor, an application-specific integrated circuit, a microprocessor, a microcontroller or any other form of programmable hardware.



    类似技术:
    公开号 | 公开日 | 专利标题
    Babic et al.2008|A review of automated feature recognition with rule-based pattern recognition
    CN100468418C|2009-03-11|Method and program for generating volume data from boundary representation data
    Lockett et al.2005|Graph-based feature recognition for injection moulding based on a mid-surface approach
    Ran et al.2010|Design of internal pins in injection mold CAD via the automatic recognition of undercut features
    Chakraborty et al.2009|Automatic determination of parting directions, parting lines and surfaces for two-piece permanent molds
    CN106959669B|2021-05-07|Notch detection
    Singh et al.2013|Systematic approach for automated determination of parting line for die-cast parts
    Doñate et al.2013|New methodology for demoldability analysis based on volume discretization algorithms
    Zhang et al.2010|Feature extraction from freeform molded parts for moldability analysis
    ES2244047T3|2005-12-01|METHOD FOR MODELING THREE-DIMENSIONAL OBJECTS AND FOR FLUID FLOW SIMULATION.
    Xu et al.2015|Machining feature recognition from in-process model of NC simulation
    ES2594772B2|2017-09-19|Method and system of obtaining virtual geometries and detection of non-moldable areas in parts
    Chen et al.2002|A region based method to automated design of multi-piece molds with application to rapid tooling
    Huang et al.2003|Generating sacrificial multi-piece molds using accessibility driven spatial partitioning
    Banerjee et al.2007|Geometrical algorithms for automated design of side actions in injection moulding of complex parts
    Mercado-Colmenero et al.2017|An automated manufacturing analysis of plastic parts using faceted surfaces
    Paramio et al.2006|A procedure for plastic parts demoldability analysis
    CN109983509A|2019-07-05|A kind of instant boolean operation method using geometric surface
    Madan et al.2007|Die-casting feature recognition for automated parting direction and parting line determination
    Verma et al.2004|Feature vector: a graph-based feature recognition methodology
    Coll Sans2014|Robust volume mesh generation for non-watertight geometries
    Li2003|Automatic parting surface determination for plastic injection mould
    Bouhadja et al.2015|Survey on simulation methods in multi-axis machining
    Lai et al.2018|Automatic recognition and decomposition of rib features in thin-shell parts for mold flow analysis
    Srikumaran et al.2005|Proving manufacturability at the design stage using commercial modeling Software: Through feature mapping and feature Accessibility
    同族专利:
    公开号 | 公开日
    ES2594772B2|2017-09-19|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US6223092B1|1990-02-14|2001-04-24|Hitachi, Ltd.|Automatic manufacturing evaluation method and system|
    WO2004042481A1|2002-10-01|2004-05-21|Agency For Science, Technology And Research|Mold design method and system|
    US20110071790A1|2009-09-24|2011-03-24|Fuji Xerox Co., Ltd.|Unmoldability determination apparatus, computer readable medium, and unmoldability determination method|
    US20110093106A1|2009-10-19|2011-04-21|Geometric Limited|Manufacturability Evaluation of Injection Molded Plastic Models Using a CAD Based DFX Evaluation System|
    法律状态:
    2017-09-19| FG2A| Definitive protection|Ref document number: 2594772 Country of ref document: ES Kind code of ref document: B2 Effective date: 20170919 |
    优先权:
    申请号 | 申请日 | 专利标题
    ES201531899A|ES2594772B2|2015-12-23|2015-12-23|Method and system of obtaining virtual geometries and detection of non-moldable areas in parts|ES201531899A| ES2594772B2|2015-12-23|2015-12-23|Method and system of obtaining virtual geometries and detection of non-moldable areas in parts|
    [返回顶部]