JP6579815B2 - Support arrangement determination apparatus, three-dimensional modeling system, and support arrangement determination method - Google Patents

Description

  The present invention relates to a support arrangement determination device, a three-dimensional modeling system, and a support arrangement determination method.

  2. Description of the Related Art Conventionally, a three-dimensional modeling apparatus that forms a desired three-dimensional modeled object (hereinafter referred to as a target modeled object) by sequentially laminating resin materials having a predetermined cross-sectional shape and curing the resin material is known. In this type of three-dimensional modeling apparatus, first, data of a cross-sectional shape of a target model is prepared using a CAD apparatus or the like. Next, using the data of the cross-sectional shape, the photocurable resin is cured, and a resin layer having a shape corresponding to the cross-sectional shape is formed. And an object model is modeled by laminating sequentially a resin layer corresponding to a section shape.

  For example, as shown in Patent Document 1, the three-dimensional modeling apparatus is a table in which an opening is formed, a tank placed on the table and containing a photocurable resin, and freely movable up and down disposed above the tank. A holder and an optical device that is disposed below the table and emits light. The light irradiated from the optical device is irradiated to the photocurable resin in the tank through the opening of the table. Of the photocurable resin stored in the tank, the portion irradiated with light is cured. By controlling the light irradiation position, the position of the resin to be cured can be changed as appropriate, and a resin layer having a desired cross-sectional shape can be formed. And a resin layer is continuously formed toward the downward direction by raising a holder one by one. In this way, a desired target model is modeled.

  By the way, when raising a holder one by one, the resin layer already shape | molded will support the load of all the resin layers modeled below the said resin layer. Therefore, for example, when there is a resin layer having a small cross-sectional area, the load of all resin layers below the resin layer may not be supported. As a result, part of the target model may be damaged during the modeling of the target model. In order to prevent such damage, as shown in FIG. 22, calculation is performed by a dedicated device such as a CAD device, so that the target modeling object 170 is formed between a part of the target modeling object 170 and the holder 113 during modeling. A plurality of support modeling objects 130 for supporting a part of the load are added, and the target modeling object 170 and the support modeling object 130 are modeled. Hereinafter, the support model is simply referred to as “support”.

JP 2003-39564 A

  By the way, the support 130 is added and arrange | positioned on the surface facing the holder 113 among the target modeling objects 170, for example. However, depending on the orientation of the target modeling object 170 with respect to the holder 113, when the target modeling object 170 is projected onto the projection surface of the target modeling object 170 with respect to the holder 113, that is, the surface of the holder 113 facing the target modeling object 170. The area of the projection surface showing the outer peripheral shape of the target object 170 to be obtained is reduced. As a result, the number of supports 130 to be added and arranged on the target model 170 is reduced, and the support 130 may not be able to sufficiently support the load of the target model 170 during modeling. Further, for example, in the thick part 170a and the thin part 170b in the target model 170, the thin part 170b is supported by the support 130 that is thinner than the thick part 170a. Can do. However, conventionally, the thickness 130 and the thickness of the support 130 to be added and arranged are determined based on the user's rule of thumb. Therefore, depending on the user, the support 170 may not be added and arranged at an optimal position.

  The present invention has been made in view of such a point, and the object thereof is to provide an optimal three-dimensional modeling apparatus that adds and arranges a support to a target modeling object that is a modeling target to model the target modeling object and the support. To provide a support arrangement determining apparatus, a three-dimensional modeling system, and a support arrangement determining method capable of adding and arranging a thickness support at an optimal position.

  The support arrangement determining apparatus according to the present invention adds and arranges a support to a target object to be modeled, and determines a support arrangement position in the three-dimensional modeling apparatus that models the target object and the support. This is an arrangement determining device. The support includes a first support having a first thickness and a second support having a second thickness that is smaller than the first thickness. The arrangement determination device includes a storage unit, a reference unit, a center of gravity calculation unit, a spindle setting unit, a rotation angle candidate generation unit, an arrangement plane determination unit, a reference point setting unit, a region dividing unit, A thickness candidate generation unit, a number calculation unit, and a determination unit are provided. The storage unit stores data of a three-dimensional model of the target object. The reference unit moves and rotates the three-dimensional model stored in the storage unit so as to have a predetermined reference position and orientation. The center of gravity calculation unit calculates the center of gravity of the three-dimensional model. The main axis setting unit calculates a farthest point farthest from the center of gravity of the three-dimensional model among points on the outer peripheral surface of the three-dimensional model, and calculates the center of gravity and the farthest point of the three-dimensional model. Set the main axis to be connected. The rotation angle candidate generator generates a plurality of rotation angle candidates for rotating the three-dimensional model around the main axis with reference to the three-dimensional model arranged so as to have the predetermined reference position and orientation. Generate. The arrangement plane determining unit determines an arrangement plane on which the support is added and arranged in each of the three-dimensional models rotated based on the plurality of generated rotation angles. The reference point setting unit draws a perpendicular line from the center of gravity to the arrangement surface, and sets a reference point where the perpendicular line and the arrangement surface intersect. The region dividing unit is a region within a predetermined range including the reference point set by the reference point setting unit on each of the arrangement planes, and a predetermined number of the first supports are added and arranged. The area is divided into a first area and a second area excluding the first area where the second support is added and arranged. The first thickness candidate generation unit creates a plurality of candidates for the first thickness that is the thickness of the first support. The number calculation unit calculates the number of the second support added and arranged in the second area on each arrangement plane determined by the arrangement plane determination unit based on a predetermined rule. The determination unit includes the plurality of rotation angles generated by the rotation angle candidate generation unit and the first thicknesses of the plurality of first supports generated by the first thickness candidate generation unit. The rotation angle at which the support can support the load of the target object and the number of the second supports is minimized, and the first thickness of the first support. decide.

  According to the arrangement determining apparatus, the first area of the arrangement surface is an area near the center of gravity among the areas of the arrangement surface. Therefore, by making the thickness of the first support added and arranged in the first region thicker than the thickness of the second support added and arranged in the second region, the support becomes the center of gravity of the target object. It is possible to support the load of the target modeled object around the center. Therefore, the support can stably support the load of the target object. In the present invention, the rotation angle is such that the support can support the load of the target object and the number of second supports added and arranged in the second region of the arrangement surface is minimized. The angle of rotation about the main axis of the three-dimensional model arranged in a predetermined reference direction and position. Therefore, the support can support the load of the target modeling object by modeling the target modeling object in the direction determined as described above. Moreover, since the number of 2nd supports can be decreased, the material required when modeling a support can be decreased.

  According to a preferred aspect of the present invention, the determination unit uses the rotation angle generated by the rotation angle candidate generation unit and the first thickness generated by the first thickness candidate generation unit as a gene. And using a genetic algorithm, the rotation angle such that the support can support the load of the target object and the number of the second supports is minimized, and The first thickness is determined.

  For example, when all possible combinations of the rotation angle and the thickness of the first support are obtained and the optimum orientation of the target object and the optimum thickness of the first support are obtained, the calculation processing becomes enormous, The load increases. However, according to the above aspect, the combination of the optimal rotation angle and the thickness of the first support is determined by applying a genetic algorithm. Therefore, by applying the genetic algorithm, it is possible to prevent the calculation time from becoming enormous as compared with the case of obtaining all possible combinations of the rotation angle and the thickness of the first support. As a result, the calculation load can be reduced.

According to another preferred aspect of the present invention, the evaluation function J used in the genetic algorithm is such that the number of the first support is p, the thickness of the first support is x, and the number of the second support is q, the thickness of the second support is β, the tensile strength of one of the supports is k, the load of the target object is Gt, and the adhesion between the target object and the three-dimensional modeling apparatus is As, If the safety factor is S,
J = (x.k.p + .beta..k.q) / S-max (Gt, As)
It is represented by The determination unit combines the plurality of rotation angles generated by the rotation angle candidate generation unit and the first thicknesses of the plurality of first supports generated by the first thickness candidate generation unit. A plurality of individuals are generated, and values obtained from the evaluation function are calculated for each of the plurality of generated individuals, so that the value of the evaluation function is greater than 0 and the number of the second supports is minimized. The individual is extracted, and the rotation angle and the first thickness of the first support are determined.

  According to the above aspect, by using the evaluation function J as described above, the target model is set in a direction that can withstand the larger load of the load Gt and the adhesion force As of the target model to be modeled. And a first support of optimum thickness can be added and placed.

  According to another preferable aspect of the present invention, the target object has a prohibited surface on which the support is not added or arranged. The said rotation angle candidate production | generation part does not make the said rotation angle which adds and arrange | positions the said support to the said prohibition surface as a candidate.

  For example, when the target modeling object is a modeling object that can be clearly identified between a surface part exposed when placed on the ground and a bottom part in contact with the ground, an arrangement surface is set in the bottom part, and the bottom part It may be preferable to add and place support only at the same time. In other words, the target object to be modeled may have a surface on which support is not desired to be added or arranged due to a design reason or the like. According to the above aspect, a surface on which support is not added and arranged is prohibited. Then, the rotation angle candidate generation unit can select only rotation angles that are oriented so as not to add and place support on the prohibited surface. Therefore, the inclination angle at which the support is not added and arranged on the prohibited surface can be set to an angle that rotates from the predetermined reference direction. Therefore, the support can support the load of the target object without adding and arranging the support on the prohibited surface.

  A three-dimensional modeling system according to the present invention includes the three-dimensional modeling apparatus and the support arrangement determination apparatus described in any one of the above-described ones.

  According to the above aspect, it is possible to provide a three-dimensional modeling system including the support arrangement determination device described in any of the above.

  The support arrangement determining method according to the present invention determines the arrangement position of the support in a three-dimensional modeling apparatus that adds and arranges a support to a target modeling object that is a modeling target and models the target modeling object and the support. This is an arrangement determination method to be performed. The support includes a first support having a first thickness and a second support having a second thickness that is smaller than the first thickness. The arrangement determination method includes a reference process, a center of gravity calculation process, a spindle setting process, a rotation angle candidate generation process, an arrangement surface determination process, a reference point setting process, a region dividing process, and a first thickness candidate generation. And a number calculation step and a determination step. In the reference step, the three-dimensional model is moved and rotated so that a predetermined reference position and orientation are obtained. In the center of gravity calculation step, the center of gravity of the three-dimensional model is calculated. In the main axis setting step, a farthest point farthest from the centroid of the three-dimensional model among points on the outer peripheral surface of the three-dimensional model is calculated, and the centroid and the farthest point of the three-dimensional model are calculated. Set the main axis to be connected. In the rotation angle candidate generation step, a plurality of rotation angle candidates for rotating the three-dimensional model around the main axis is generated with reference to the three-dimensional model arranged so as to have the predetermined reference position and orientation. To do. In the arrangement plane determination step, an arrangement plane on which the support is added and arranged is determined in each of the three-dimensional models rotated based on the plurality of generated rotation angles. In the reference point setting step, a perpendicular line from the center of gravity to the arrangement surface is drawn, and a reference point at which the perpendicular line intersects the arrangement surface is set. In the region dividing step, each of the arrangement planes is a region within a predetermined range including the reference point set in the reference point setting step, and a predetermined number of the first supports are added and arranged. The first area is divided into an area excluding the first area and a second area where the second support is added and arranged. In the first thickness candidate generation step, a plurality of first thickness candidates that are the thicknesses of the first support are created. In the number calculating step, the number of the second supports added and arranged in the second region is calculated based on a predetermined rule in each of the arrangement surfaces determined in the arrangement surface determination step. In the determination step, among the plurality of rotation angles generated in the rotation angle candidate generation step and the first thicknesses of the plurality of first supports generated in the first thickness candidate generation step, The rotation angle and the first thickness of the first support are determined such that the support can support the load of the target object and the number of the second supports is minimized. .

  According to another preferable aspect of the present invention, in the determination step, the rotation angle generated in the rotation angle candidate generation step and the first thickness generated in the first thickness candidate generation step are genes. And using a genetic algorithm, the rotation angle such that the support can support the load of the target object and the number of the second supports is minimized, and The first thickness is determined.

According to another preferred aspect of the present invention, the evaluation function J used in the genetic algorithm is such that the number of the first support is p, the thickness of the first support is x, and the number of the second support is q, the thickness of the second support is β, the tensile strength of one of the supports is k, the load of the target object is Gt, and the adhesion between the target object and the three-dimensional modeling apparatus is As, If the safety factor is S,
J = (x.k.p + .beta..k.q) / S-max (Gt, As)
It is represented by In the determination step, a plurality of the rotation angles generated in the rotation angle candidate generation step and a plurality of the first thicknesses of the first support generated in the first thickness candidate generation step Generating an individual, calculating a value obtained from each of the generated evaluation functions of the plurality of individuals, wherein the value of the evaluation function is greater than 0, and the number of the second supports is minimized An individual is extracted, and the rotation angle and the first thickness of the first support are determined.

  According to another preferable aspect of the present invention, the target object has a prohibited surface on which the support is not added or arranged. In the rotation angle candidate generation step, the rotation angle that adds and arranges the support on the prohibited surface is not a candidate.

  According to the present invention, it is possible to add and arrange a support having an optimum thickness for an object to be shaped at an optimum position.

It is sectional drawing of the three-dimensional modeling system which concerns on embodiment of this invention. It is a top view of a three-dimensional modeling system. It is the schematic diagram which showed the state which models the whole modeling thing in a holder. It is sectional drawing in the IV-IV cross section of FIG. It is a perspective view which shows an example which added and arrange | positioned the support to the target molded article. It is a perspective view which shows an example which added and arrange | positioned the support to the target molded article. It is explanatory drawing of the main axis | shaft of a target molded article. It is a figure explaining dividing | segmenting an arrangement surface into a 1st area | region and a 2nd area | region, and is a front view of a target molded article. It is a block diagram of an arrangement determining device. It is the flowchart which showed the procedure which determines what support is added and arrange | positioned on which surface of a target model object. It is the flowchart which showed the procedure which sets a target molded article model to a reference | standard position and direction. It is the perspective view which showed an example of the target model object. It is the perspective view which showed an example of the target model object after inclining. It is the perspective view which showed an example of the target model object after making it rotate. It is a front view of the object model object which showed an example of the state which added and arranged support to the object model object model. It is a bottom view of the object model object which showed an example of the state which added and arranged support to an object model object model. It is a figure which shows the individual of an initial population. It is a figure which shows the individual after a sort. It is a figure which shows an example in which a new individual is produced | generated by two-point crossover. 20A to 20C are diagrams illustrating an example in which a new individual is generated by mutation. It is a figure which shows a prohibition surface, and is a perspective view which shows an example of a target molded article. In the prior art, it is the schematic diagram which showed the state which models the target molded article support which added the support to the holder.

  Hereinafter, a three-dimensional modeling system including a support arrangement determination device (hereinafter referred to as an arrangement determination device) and a support arrangement determination method according to an embodiment of the present invention will be described with reference to the drawings. It should be noted that the embodiments described herein are not intended to limit the present invention. In addition, members / parts having the same action are denoted by the same reference numerals, and overlapping descriptions are omitted or simplified as appropriate.

<First Embodiment>
FIG. 1 is a cross-sectional view of a three-dimensional modeling system 10 according to the present embodiment. FIG. 2 is a plan view of the three-dimensional modeling system 10. Note that symbols F, Rr, L, and R in the drawings indicate front, rear, left, and right, respectively. However, these are only directions for convenience of explanation, and do not limit the installation mode of the three-dimensional modeling system 10 at all.

  The three-dimensional modeling system 10 is a system that can model a target model using a support. Hereinafter, the entire target model and the support are referred to as “total model”. In the three-dimensional modeling system 10, data of a plurality of cross-sectional shapes of the entire modeled object is used. Here, the “cross-sectional shape” is a cross-sectional shape of each layer when the entire shaped article is divided into a plurality of layers. In the three-dimensional modeling system 10, a liquid photocurable resin is cured to model a resin layer having a shape corresponding to the cross-sectional shape. And the whole modeling thing is modeled by laminating | stacking the resin layer sequentially. The “photocurable resin” is a resin that is cured when irradiated with light having a predetermined wavelength.

  As shown in FIG. 1, the 3D modeling system 10 determines the orientation of the target modeled object and forms the shape of the support before modeling the entire modeled object in the 3D modeling apparatus 10A and the 3D modeling apparatus 10A. And an arrangement determining apparatus 100 (see FIG. 9) for determining an arrangement position.

  The three-dimensional modeling apparatus 10 </ b> A includes a table 11, a tank 12, a holder 13, an optical device 14, and a control device 16.

  The base 11 is supported by the case 25. An opening 21 through which light passes is formed in the table 11. The tank 12 contains a liquid photocurable resin 23. The tank 12 is mounted on the table 11 so as to be attachable. As shown in FIG. 2, the tank 12 covers the opening 21 of the table 11 when placed on the table 11. The tank 12 is good to be formed with the material which can permeate | transmit light, for example, a transparent material.

  As shown in FIG. 1, the holder 13 is disposed above the tank 12 and above the opening 21 of the table 11. The holder 13 is configured to be movable up and down. The holder 13 is immersed in the photocurable resin 23 in the tank 12 when it is lowered, and when it is raised, the holder 13 lifts the photocurable resin 23 that is cured by being irradiated with light. Here, the column 11 is provided with a support column 41 extending in the vertical direction. A slider 42 is attached in front of the column 41. The slider 42 can freely move up and down along the support column 41, and moves upward or downward by the motor 43. Here, the holder 13 is attached to the slider 42. The holder 13 is moved upward or downward by the motor 43.

  The optical device 14 is disposed below the table 11. The optical device 14 is a device that irradiates the liquid photocurable resin 23 contained in the tank 12 with light having a predetermined wavelength. The optical device 14 is accommodated in a case 25 provided below the table 11. The optical device 14 includes a projector 31 and a mirror 32. The projector 31 is a light source that emits light. The mirror 32 is a member that reflects the light emitted from the projector 31 toward the tank 12. The mirror 32 is disposed below the opening 21 formed in the base 11 and behind the projector 31. The light emitted from the projector 31 is reflected by the mirror 32 and is applied to the photocurable resin 23 in the tank 12 through the opening 21 of the table 11. However, the arrangement and configuration of the optical device 14 are not particularly limited.

  The control device 16 is connected to a motor 43 that controls the slider 42 to which the holder 13 is attached to freely move up and down, and a projector 31 of the optical device 14. The control device 16 drives the motor 43 to move the slider 42 and the holder 13 upward or downward. Further, the control device 16 controls the energy, light intensity, light amount, light wavelength band, light shape, light irradiation position, light emission timing, and the like emitted from the projector 31. The configuration of the control device 16 is not particularly limited. For example, the control device 16 is a computer, and may include a central processing unit (hereinafter referred to as a CPU), a ROM storing a program executed by the CPU, a RAM, and the like.

  The above is the configuration of the three-dimensional modeling apparatus 10A. As described above, the three-dimensional modeling apparatus 10A models a target modeled object to which support is added and arranged. Next, support will be described.

  When the three-dimensional modeling apparatus 10A models the target modeled object, the holder 13 is sequentially raised every time the resin layer is modeled, and a new resin layer is modeled below the resin layer. However, for example, after a resin layer having a large cross-sectional area is formed, a relatively small resin layer, for example, a resin layer having a small size such that the contact area with the holder 13 is less than half of the large resin layer is present. In this case, the resin layer may not be able to support the load of the entire resin layer including the large resin layer positioned below the resin layer. As a result, there is a possibility that the target object is damaged during the modeling. Therefore, a support is added and arranged with respect to the target model so that the load of the target model during the modeling can be sufficiently supported. Thereby, it can prevent that a target model is damaged during modeling.

  For example, if the three-dimensional modeling apparatus 10 </ b> A directly models the target model on the holder 13, the target model is peeled off from the holder 13 after modeling. In that case, when peeling the part which is contacting the holder 13 among the object modeling objects from the holder 13, the object modeling object may be damaged. Therefore, as shown in FIG. 3, a support 30 is added and arranged between the holder 13 and the target object 70a. Then, after the modeling is completed, the entire modeled object (modeled object in which the support 30 is added and arranged on the target modeled object 70a) is peeled off from the holder 13, and then the support 30 may be removed from the target modeled object 70a. Thereby, it is possible to prevent the target object 70a from being damaged.

  The shape of the support 30 is not particularly limited. 4 is a cross-sectional view taken along the line IV-IV in FIG. In the present embodiment, as shown in FIG. 4, a plurality of columns are used as the support 30. Here, as shown in FIG. 3, the upper end part of the support 30 is provided in the holder 13, and the lower end part is provided in the target object 70a. However, the shape of the support 30 is merely an example. The shape of the support 30 may be, for example, a triangular or quadrangular cross-sectional shape. The shapes of the plurality of supports 30 may be the same shape or may be partially different shapes. The interval between the adjacent supports 30 is not particularly limited, but is constant here as shown in FIG. In the present embodiment, the plurality of supports 30 are added and arranged at equal intervals. The plurality of supports 30 are added and arranged at positions aligned in the left-right direction and the front-rear direction. However, the arrangement position of the support 30 is merely an example. The interval between the adjacent supports 30 may not be constant. The plurality of supports 30 may be added and arranged in a zigzag manner, for example.

  Next, conditions for the support 30 to support the load of the target object will be described. First, in order for the support 30 to support the load of the target object, the load that can be supported by the support 30 must be larger than the load of the target object. Further, in the three-dimensional modeling apparatus 10 </ b> A, an adhesion force is generated between the surface of the tank 12 and the resin layer. Therefore, in order to peel the resin layer from the tank 12, the target modeled object is larger than the adhesion force. Need to lift. At that time, a force having the same magnitude as the adhesion force is applied to the support 30. Therefore, in order for the support 30 not to be damaged, the load that can be supported by the support 30 must be larger than the above-mentioned adhesion force. As described above, the load that can be supported by the support 30 (hereinafter referred to as a support load) is Ls [gf], the load of the target object is Gt [gf], and the adhesion between the resin layer of the target object and the tank 12 is as follows. (Here, the adhesion force is a force that combines the load of the resin layer and the adhesion force generated between the surface of the tank 12 and the resin layer, as will be described later.) As [gf], Then, Ls must be larger than the larger of Gt and As. That is, Ls> max (Gt, As) must be satisfied.

The support load Ls can be expressed by the following formula (1), where Sp is the area of the one support 30 that contacts the target object, Sp is the tensile strength of one support 30, and I is the number of supports 30. .
Ls = Sp × Kp × I (1)

The load Gt of the target model can be expressed by the following formula (2), where Va is the volume of the target model and Da is the density of the target model.
Gt = Va × Da (2)

The adhesion force As is defined as follows: Vn is the volume of the target model when layers 1 to n are stacked, Sn is the cross-sectional area of the target model in the nth layer, and K is the tensile coefficient (tensile stress) of the target model. And can be represented by the following formula (3).
As = Vn × Da + Sn × K (3)

By the way, logically, if Ls> max (Gt, As), the support 30 is not damaged. However, in practice, the support 30 may be damaged due to various uncertain factors even if Ls> max (Gt, As). Therefore, a safety factor S composed of a real number larger than 1 may be introduced. When the safety factor S is introduced, a condition for the support 30 to sufficiently support the load of the object to be modeled may be set as in the following formula (4).
Ls / S> max (Gt, As) (4)
For example, when the safety factor S = 3, the support 30 can withstand a force up to three times the larger one of the load Gt and the adhesion force As of the target modeled object. Therefore, by satisfy | filling said Formula (4), the support 30 can fully support the load of a target molded article.

  As described above, the three-dimensional modeling apparatus 10 </ b> A models the target model by sequentially raising the holder 13. The target object is modeled in order from the upper part to the lower part. The three-dimensional modeling apparatus 10A receives data of a three-dimensional model (hereinafter, referred to as a target modeling model) of a target modeling object, and uses the data to form a target modeling object having substantially the same shape as the target modeling model. Is shaped.

  The target model object has a predetermined orientation (hereinafter referred to as an initial orientation). This initial orientation is usually determined according to the data creation procedure of the target object model. For example, when the target model object is a conical ornament, the initial orientation is determined so that the apex of the cone is on the top and the bottom is on the bottom. However, as will be described below, it is not always appropriate to form the target object in the initial orientation.

  Depending on the orientation of the target model, the area of the surface of the target model facing the holder 13 may be small, and a sufficient number of supports 30 may not be added and arranged. For example, when the support 30 is added and arranged in the orientation of the target object 70b as shown in FIG. 5, the area of the surface 71b1 of the target object 70b to which the support 30 is added and arranged is small. Therefore, the number of supports 30 that can be added is reduced. As a result, during modeling, the support 30 may not be able to support the load of the target model 70b due to the load of the target model 70b. However, when the support 30 is added and arranged in the orientation of the target object 70b as shown in FIG. 6, the area of the surface 71b2 of the target object 70b to which the support 30 is added and arranged is large. Therefore, the number of supports 30 to be added and arranged is larger than that in the example of FIG. As a result, the force with which the support 30 supports the load of the target object 70b is increased. As shown in FIG. 6, the applicant of the present application sufficiently supports the load of the target object 70 b when the area of the surface of the target object 70 b to which the support 30 can be added can be sufficiently secured. I found that I can do it.

  Therefore, the applicant of the present application paid attention to the main shaft 78c of the target object 70c as shown in FIG. Here, the “principal axis” refers to the center of gravity 74c of the target object 70c and the point 76c farthest from the center of gravity 74c among the points on the outer peripheral surface of the target object 70c (hereinafter referred to as the farthest point 76c). A connecting line. A surface parallel to the main axis 78c of the target object 70c or a surface 77 having a small inclination angle with respect to the main axis 78c can sufficiently secure an area as compared with other surfaces. Therefore, for example, the support 30 can support the load of the target object 70c with a sufficient support load by setting the surface 77 as an arrangement surface on which the support 30 is added and arranged. In addition, the detailed procedure which sets the gravity center and main axis | shaft of a target molded article is mentioned later.

  By the way, if the number of the supports 30 is too small, the load of the target object cannot be supported with a sufficient support load. On the other hand, the support 30 is a member that is removed from the target model after the modeling is completed. Therefore, if the number of supports 30 is large and the contact area per unit area between the support 30 and the target object is too large, it takes much time and effort to remove the support 30 from the target object. Moreover, useless photocurable resin 23 increases. From the above viewpoint, it is preferable to reduce the number of supports 30 and reduce the contact area per unit area between the support 30 and the target object. Therefore, the applicant of the present application assumes that the support 30 is capable of supporting the load of the target object with a sufficient support load, that is, assuming that the above expression (4) is satisfied. It was examined how the amount of the photocurable resin 23 used for the support 30 can be reduced by adding and arranging the support 30 on the surface.

  Here, the applicant of the present application paid attention to the center of gravity 74c of the target object 70c as shown in FIG. When the support 30 is added and arranged directly below the center of gravity 74c of the target object 70c, the bending moment becomes “0”. Therefore, the support 30 supports the load of the target object 70c by adding and arranging the support 30 around the region 77a located near the center of gravity 74c of the target object 70c in the surface of the target object 70c. I found it easy to do. In addition, most of the load of the target object 70c is supported by the support 30 added to the region 77a located near the center of gravity 74c. In this case, the thickness of the support 30 to be added is larger in the region 77b located at a position away from the center of gravity 74c of the target model 70c than in the region 77a located near the center of gravity 74c. It has been found that the support 30 can stably support the load of the target object 70c even if the thickness is reduced.

  Therefore, for example, as illustrated in FIG. 8, the surface 82 d (hereinafter, referred to as a placement surface 82 d) of the target model 70 d to which the support 30 is added and arranged is positioned near the center of gravity 74 d of the target model 70 d. The region is defined as a first region 91d, and the region excluding the first region 91d in the arrangement surface 82d is defined as a second region 92d. The second region 92d is a region away from the center of gravity 74d of the target object 70d as compared to the first region 91d. Here, the support 30 added and arranged in the first area 91d is referred to as a first support 30L. The support 30 added and arranged in the second area 92d is referred to as a second support 30S. The first support 30L and the second support 30S are added and arranged at a constant equal interval, and the thickness (first thickness) of the first support 30L is set to be the thickness (first thickness) of the second support 30S. (Thickness of 2). As a result, the support 30 can stably support the load of the target model 70d around the center of gravity 74d of the target model 70d.

In addition, when the thickness of the support 30 to be added and arranged is different between the first region and the second region, the support load Ls can be expressed by the following formula (5).
Ls = x · k · p + β · k · q (5)
In the above formula (5), x represents the thickness of the first support 30L, k represents the tensile strength of one support 30, and p represents the number of the first supports 30L. β represents the thickness of the second support 30S, and q represents the number of the second supports 30S. The thickness x of the first support 30L can be expressed by x = β · c, where c (hereinafter also referred to as a first thickness coefficient c) is the ratio of the second support 30S to the thickness β. .

  Conventionally, the determination as to which surface of the target object is to be modeled and the correspondence of the support 30 is often made based on the user's rule of thumb. Therefore, depending on the user's decision, the support 30 may not be able to support the load of the target object due to the load of the target object. Therefore, in the present embodiment, the arrangement determining device 100 determines the thickness and the arrangement position of the support 30 so that the support 30 can sufficiently support the load of the target object.

  FIG. 9 is a block diagram of the arrangement determining apparatus 100. The arrangement determining apparatus 100 may be separate from the 3D modeling apparatus 10A, or may be incorporated in the 3D modeling apparatus 10A. For example, the arrangement determining apparatus 100 is a computer, and may include a CPU, a ROM storing a program executed by the CPU, a RAM, and the like. Here, the arrangement position of the support 30 is determined using a program stored in the computer. The arrangement determining apparatus 100 may be a dedicated computer for the three-dimensional modeling system 10 or a general-purpose computer.

  The arrangement determining apparatus 100 includes a storage unit 52, a preprocessing unit 54, a centroid calculation unit 56, an initial unit 57, a spindle setting unit 58, a reference unit 60, a rotation angle candidate generation unit 62, and a first thickening unit. A candidate generation unit 63, an arrangement plane determination unit 64, a reference point setting unit 65, a region division unit 66, a number calculation unit 67, and a determination unit 68. In addition, each part mentioned above may be comprised by software, and may be comprised by hardware. A detailed description of each part will be given according to the flowchart described later.

  FIG. 10 is a flowchart showing a procedure for determining what support 30 is added and arranged on which surface of the target object model 72. FIG. 11 is a flowchart showing a procedure for setting the target object model 72 to the reference position and orientation. 12 to 16 are perspective views of the target object model 72. Here, using the target object model 72 in FIGS. 12 to 16, what support 30 is added and arranged on which surface in the target object model 72 which is a three-dimensional model of the object object 70. The determination procedure will be described with reference to the flowcharts of FIGS. 10 and 11.

  Here, it is assumed that the support 30 is added and arranged on the lower surface of the target object model 72. However, the position at which the support 30 is added and arranged is not particularly limited. For example, the support 30 may be added and arranged on the upper surface of the target object 70.

  In the present embodiment, it is assumed that data of the target object model 72 corresponding to the target object 70 is stored in the storage unit 52 of the arrangement determining apparatus 100 in advance. The data of the target model object 72 is read into the storage unit 52 from a recording medium or another computer (not shown), for example, by a user operation. The data of the target model object 72 is, for example, data specified by a set of a plurality of points in an XYZ orthogonal coordinate system, and is three-dimensional data. In the target model object 72, the target model 70 is reproduced by combining a plurality of triangular polygons or triangular pyramid polygons.

  First, in step S <b> 101, the preprocessing unit 54 performs preprocessing on the target object model 72 stored in the storage unit 52. The target model object 72 is often detailed data that faithfully reproduces the three-dimensional shape of the target model 70. If the target model object 72 is used as it is, the arrangement determination process in the arrangement determination apparatus 100 may take a long time. Therefore, it is preferable that the preprocessing unit 54 performs preprocessing on the target model object 72 to reduce the data amount of the target model model 72. Here, the preprocessing unit 54 performs a smoothing process on the target object model 72. For example, as the smoothing process, a process of reducing the amount of data by reducing the number of triangular polygons constituting the target model object 72 may be performed. The smoothing method is not particularly limited, and for example, a conventionally known smoothing method can be applied. For example, a Gaussian function may be used as a smoothing method. Here, the data of the target object model 72 that has been subjected to the smoothing process is stored in the storage unit 52. Note that the pre-processing in step S101 can be omitted.

  Next, in step S <b> 102, the position and orientation serving as a reference for the target model object 72 are set. In the present embodiment, the initial position and orientation of the target object model 72 are set. The position and orientation of the target object model 72 are specified by three-axis orthogonal coordinates. However, it is also possible to specify the position and orientation of the target object model 72 using another coordinate system. Here, the reference position and orientation are set from the initial position and orientation of the target object model 72. The procedure of setting the target object model 72 to the reference position and orientation is performed along steps S201 to S205 in FIG.

  First, in step S <b> 201, the center of gravity calculation unit 56 calculates the center of gravity of the target model object 72. Here, a conventionally known method can be used as a method of calculating the center of gravity of the target object model 72. For example, the center of gravity of the target object model 72 can be calculated by using the center of gravity of each triangular pyramid polygon constituting the target object model 72. In FIG. 12, the center of gravity of the target model object 72 is a point 74.

For example, the center of gravity 74 of the target object model 72 can be calculated by the center of gravity calculation unit 56 as follows. For example, it is assumed that the target object model 72 is configured by combining a plurality of triangular pyramid polygons A1, A2,. Here, first, the center-of-gravity calculation unit 56 calculates the volume Vall of the target object model 72. Next, the center-of-gravity calculation unit 56 calculates the polygonal pyramid polygons A1, A2,..., An gravity centers G1, G2,..., Gn and the triangular pyramid polygons A1, A2,. Volumes V1, V2,..., Vn are calculated. At this time, the center of gravity 74 of the target object model 72 is calculated by the following equation (6).
Center of gravity 74 = (V1 × G1 + V2 × G2 +... + Vn × Gn) / Vall (6)

  Next, in step S <b> 202, the initial unit 57 sets the initial position and orientation of the target object model 72. Specifically, the initial unit 57 sets the initial position so that the center of gravity 74 of the target object model 72 calculated in step S201 is the center position of the holder 13 (see FIG. 1) in plan view. The initial orientation of the target model object 72 is, for example, the direction when the target model model 72 is created. Note that the initial position and orientation of the target model object 72 are not particularly limited. Here, the initial part 57 moves and rotates the target model object 72 so as to be in the initial position and orientation. The target object model 72 set to the initial position and orientation is stored in the storage unit 52.

  Next, in step S <b> 203, the main axis setting unit 58 sets the main axis of the target model object 72. As described above, the “principal axis” is a triangular polygon farthest from the center of gravity 74 out of the center of gravity 74 of the object model 72 and the center of the triangle polygon of the point on the outer peripheral surface of the object model 72. The line connecting the center of gravity (hereinafter referred to as the farthest point). For example, the main axis of the target model object 72 can be set by the following procedure. First, the main axis setting unit 58 calculates the distance between the center of gravity 74 of the target object model 72 and the center of gravity of the polygons of each triangle that constitutes the target object model 72. Here, the center of gravity of the polygon calculated in step S201 can be used as the center of gravity of each triangular polygon constituting the target model object 72. The main axis setting unit 58 obtains the center of gravity of the triangular polygon corresponding to the longest distance among the distances calculated above. In the target object model 72 of FIG. 12, the triangular polygon farthest from the center of gravity 74 of the target object model 72 is the polygon 76a, and the farthest point is the center of gravity 76 of the polygon 76a. Here, the main axis setting unit 58 sets the main axis 78 as a straight line through which the center of gravity 74 and the farthest point 76 of the target object model 72 pass. Strictly speaking, the center of gravity 76 of the polygon 76 a is not necessarily the farthest point farthest from the center of gravity of the target object model 72. This is because the polygon 76 a is a plane having a certain area centered on the center of gravity 76. However, since the polygon 76a is usually a sufficiently small area with respect to the surface area of the target object model 72, the center of gravity 76 can be regarded as the farthest point.

  Next, in step S204, as shown in FIG. 13, the reference portion 60 is set to the initial position and orientation by the initial portion 57 so that the main axis 78 of the target object model 72 is parallel to the horizontal plane H. The target object model 72 is tilted. Here, the horizontal plane H is the surface of the holder 13 on which the target model 70 is modeled, that is, the lower surface of the holder 13 (the surface facing the target model 70). Therefore, the reference unit 60 tilts the target object model 72 set to the initial position and orientation so that the main axis 78 of the target object model 72 is parallel to the lower surface of the holder 13. Note that the target object model 72 after being tilted is stored in the storage unit 52.

  Next, in step S <b> 205, the reference unit 60 rotates the target model object 72. Specifically, first, the reference unit 60 determines the center of gravity of the target object model 72 among the center of gravity of the triangular polygons on the outer peripheral surface of the target object model 72 with respect to the target object model 72 tilted in step S204. The nearest point which is the closest point from 74 is obtained. In FIG. 13, the closest point is a point 80 located on the upper surface of the target object model 72. Then, the reference unit 60 is a target modeling with the main shaft 78 as an axis so that the closest point 80 overlaps with the main shaft 78 in a plan view and is positioned below the main shaft 78 as shown in FIG. The object model 72 is rotated. Here, the reference unit 60 rotates the target object model 72 about the main axis 78 so that the closest point 80 is the closest position to the horizontal plane H, that is, the surface of the holder 13. The closest point 80 is located on the lower surface of the target model object 72 after being rotated. In the present embodiment, the position and orientation of the target object model 72 after being rotated by the reference portion 60 become the reference position and orientation. The object model 72 after being rotated is stored in the storage unit 52.

  As described above, the steps after step S103 in FIG. 10 are performed on the target object model 72 set at the reference position and orientation. In step S103 and subsequent steps, with respect to the target object model 72 set to the reference position and orientation, the arrangement surface on which the support 30 is added and arranged, the thickness of the first support 30L, and the number of second supports 30S are set. decide. It is assumed that the number of first supports 30L and the thickness of the second support 30S are determined in advance. Here, the number of second supports 30S can be obtained uniformly according to a predetermined rule.

  The arrangement surface on which the support 30 is added and arranged is the lower surface of the target object model 72 after being rotated around the main axis 78 with respect to the target object model 72 set to the reference position and orientation. At this time, there are a plurality of angles (hereinafter referred to as rotation angles θ) that allow the object model 72 to be rotated from the reference position and orientation with the main shaft 78 as an axis. There are a plurality of values that the thickness of the first support 30L can take. Therefore, the number of combinations that the rotation angle θ and the thickness of the first support 30L can take is enormous, and enormous processing time is required. Therefore, in the present embodiment, a combination of the optimal rotation angle θ and the thickness of the first support 30L is determined by applying a genetic algorithm. Here, the optimal position and orientation of the target object model 72 are determined by applying a genetic algorithm, and the optimal thickness and the optimal for the target object model 72 having the determined optimal position and orientation are determined. Add and deploy a large number of supports 30.

  First, in step S103 of FIG. 10, an initial population is generated. In step S103, first, a predetermined number of individuals (hereinafter, the number of individuals is assumed to be n) is generated. For example, the thickness of the first support 30L can be obtained by the ratio to the thickness of the second support 30S, that is, the thickness of the second support 30S × the first thickness coefficient c. Each individual uses the rotation angle θ rotated from the reference position and orientation around the main axis 78 of the target object model 72 and the first thickness coefficient c as genes, and the rotation angle θ and the first thickness. The coefficients c are arranged in order. Here, the rotation angle candidate generation unit 62 has a rotation angle θ that rotates the target object model 72 about the main axis 78 with the target object model 72 arranged so as to be in the reference position and orientation as a reference. Create multiple candidates. The first thickness candidate generation unit 63 creates a plurality of candidates for the first thickness coefficient c used when determining the thickness of the first support 30L. Then, by generating each individual from the plurality of generated rotation angles θ and the plurality of first thickness coefficients c, an initial population is generated.

  By the way, in step S205 of FIG. 11, the lower surface of the target object model 72 after being rotated by the reference unit 60 is a surface close to the center of gravity 74 (see FIG. 14). Therefore, when the support 30 is added and arranged on the lower surface of the target object model 72 after being rotated, the support 30 can support the load of the object object 70 more stably. Therefore, it is preferable to add and arrange the support 30 on at least a part of the lower surface of the target object model 72 after being rotated in step S205. Therefore, the range that the rotation angle θ can take is even after a part of the lower surface of the target object model 72 set to the reference position and orientation rotates the target object model 72 according to the rotation angle θ. The range is preferably located on the lower surface. Although the range of the rotation angle θ varies depending on the target model object 72, for example, the range of the rotation angle θ is −90 ° ≦ θ ≦ 90 °. Here, the rotation angle θ among the genes of each individual is determined by the rotation angle candidate generation unit 62 using a random number within the above range.

  The range of the first thickness coefficient c may be set in advance. The lower limit value of the first thickness coefficient c is preferably larger than 1. Here, if the upper limit value of the thickness of the first support 30L is xMax and the thickness of the second support 30S is β, the thickness β of the second support 30S is expressed by β <xMax. The upper limit value of the first thickness coefficient c is not particularly limited. For example, the upper limit value of the first thickness coefficient c is preferably smaller than xMax / β. That is, the range of the first thickness coefficient c is, for example, 1 <c <xMax / β. Here, among the genes of each individual, the first thickness coefficient c is determined by the first thickness candidate generation unit 63 with a random number in 1 <c <xMax / β.

As described above, for example, n individuals G0 ₁ , G0 ₂ , G0 ₃ ,..., G0 _n as shown in FIG. In FIG. 17 to FIG. 20, the numerical values in parentheses indicate the rotation angle θ on the left and the first thickness coefficient c on the right.

  Next, in step S104, fitness is evaluated. Here, first, the arrangement surface determination unit 64 changes the orientation of the target object model 72 by rotating the target object model 72 around the main axis 78 in accordance with the rotation angle θ that is the gene of each individual. And the arrangement | positioning surface determination part 64 sets the arrangement | positioning surface which adds and arrange | positions the support 30 with respect to the target model object 72 which changed direction according to rotation angle (theta). As described above, as shown in FIG. 14, the arrangement surface is the lower surface 82 of the target object model 72 after the direction is changed according to the rotation angle θ.

  Next, the area dividing unit 66 divides the arrangement surface 82 into a first area 91 and a second area 92 as shown in FIG. For example, the arrangement surface 82 may be divided into a first area 91 and a second area 92 as follows. First, the reference point setting unit 65 sets an area reference point 86 as shown in FIG. The “region reference point” is a point that is located on the arrangement surface 82 and serves as a reference used when dividing the arrangement surface 82. In the present embodiment, the “region reference point” corresponds to the “reference point” of the present invention. The area reference point 86 can be set as follows. First, the reference point setting unit 65 draws a perpendicular 84 from the center of gravity 74 of the target object model 72 calculated by the center of gravity calculating unit 56 toward the arrangement surface 82. The perpendicular 84 is a line orthogonal to the arrangement surface 82. Then, the reference point setting unit 65 sets the intersection of the perpendicular 84 and the arrangement surface 82 as the region reference point 86. Here, the region reference point 86 is a point that overlaps with the center of gravity 74 of the target object model 72 in the bottom view. The area reference point 86 is a point closest to the center of gravity 74 among the points on the arrangement surface 82. In the present embodiment, the region reference point 86 coincides with the nearest point 80.

  Next, the area dividing unit 66 divides the arrangement surface 82. Here, the region dividing unit 66 divides the arrangement surface 82 into a first region 91 and a second region 92 with the region reference point 86 set by the reference point setting unit 65 as a reference. The first area 91 is an area close to the center of gravity 74 of the target object model 72. The first area 91 is an area within a predetermined range including the area reference point 86. The predetermined range including the area reference point 86 is a range stored in advance in the storage unit 52. The predetermined range including the area reference point 86 is not particularly limited, and is, for example, a range that is separated from the area reference point 86 by a predetermined distance. For example, the predetermined distance is a distance including a load corresponding to a predetermined ratio of the load of the target object model 72 with the region reference point 86 as the center. This predetermined distance is a numerical value predetermined in the storage unit 52. Here, the region reference point 86 is located at the center of the first region 91 as shown in FIG. The center of the first region 91 is a point located in the middle of the region. The center of the first region 91 may be obtained using a conventionally known method. The position of the area reference point 86 in the first area 91 is not particularly limited. For example, the area reference point 86 may be located at the end of the first area 91.

  The second area 92 is an area excluding the first area 91 on the arrangement surface 82. The second region 92 is a region at a position farther from the first region 91 with reference to the center of gravity 74 of the target object model 72.

As described above, after dividing the arrangement surface 82 into the first area 91 and the second area 92 by the area dividing section 66, the number calculating section 67 sets the second support 30S to the second area according to a predetermined rule. The number of second supports 30S added and arranged in the region 92 and added and arranged in the target model object 72 is calculated. Here, as described above, the plurality of second supports 30S are added and arranged in the second region 92 so as to be equally spaced in the front-rear direction and the left-right direction. Thereafter, the determination unit 68 calculates the value of the evaluation function J for each individual using the number of the second support 30S and the first thickness coefficient c which is a gene. Here, the evaluation function J can be expressed as the following formula (7) using the above formulas (1) to (5).
J = Ls / S-max (Gt, As)
= (Β · c · k · p + β · k · q) / S-max (Va · Da, Vn · Da + Sn · K) (7)

  As described above, Ls is a load that can be supported by the support 30 (support load), S is a safety factor, Gt is the load of the target object 70, and As is between the resin layer of the target object 70 and the tank 12. It shows the adhesion strength. β represents the thickness of the second support 30S, c represents the first thickness coefficient, k represents the tensile strength of one support 30, and p represents the number of the first supports 30L. q represents the number of second supports 30S. Here, β · c represents the thickness of the first support 30L. Further, Va is the volume of the target model 70, Da is the density of the target model 70, Vn is the volume of the target model 70 when the first to nth layers are stacked, and Sn is the nth layer of the target model 70. The cross-sectional area, K, represents the stress (tensile stress) of the target object 70.

Thereafter, the determination unit 68 compares the values of the evaluation functions J of the calculated n individuals G0 ₁ , G0 ₂ , G0 ₃ ,..., G0 _n . Here, the determination unit 68 regards the superior individual as the number of the second support 30S among the individuals whose evaluation function J is larger than 0, and gives a high evaluation. Then, the determination unit 68 rearranges the n individuals so that the individual who is given high evaluation is ranked higher. In this embodiment, the genetic algorithm is set so that the evaluation function J is the number of the second support 30S itself. Therefore, assuming that the n individuals after the rearrangement are G1 ₁ , G1 ₂ , G1 ₃ ,..., G1 _n in order, the evaluation functions J ₁ , J ₂ , J ₃ ,. .., J _n is as shown in FIG.

  Next, in step S105, the determination unit 68 performs selection. Selection is a step of selecting n excellent individuals from the group. In other words, the selection is to select n individuals whose evaluation is high in the fitness evaluation. In the first selection, since the number of individuals included in the group is n, all individuals are selected.

Next, in step S106, the determination unit 68 performs crossover. The crossover method is not particularly limited, and various known methods can be used. A combination of known crossover techniques may be applied. Here, two-point crossover is used. The two-point crossover, most evaluation was high individuals in the evaluation of fitness of the step S104 (hereinafter, referred to as best individuals.) G1 ₁ and, other individuals _{G1 2,} G1 3, · · ·, of the G1 _n One is paired. Then, for each pair, the gene of the rotation angle θ of the best individuals G1 _1, new individuals combining the gene of the first thickness coefficient c of other individuals, and, for Best individual G1 ₁ first A new individual is generated by combining the gene of the thickness coefficient c and the gene of the rotation angle θ of another individual. For example, as shown in FIG. 19, a new individual G1 ₁₂ and an individual G1 ₂₁ are generated from the pair of the best individual G1 ₁ and the individual G1 ₂ .

Next, in step S107, the determination unit 68 performs a mutation. In the mutation, a new individual having a new gene combination is generated according to a predetermined rule. In addition, the method of mutation is not limited at all, and various known methods can be used. Here, a predetermined number of individuals are extracted at random from individuals whose evaluation function J is 0 or less. Then, for the extracted individual, the value of the rotation angle θ or the first thickness coefficient c is replaced with a random number in the above-described range. For example, for the extracted individual, any one of the rotation angle θ, the first thickness coefficient c, and the rotation angle θ and the first thickness coefficient c is selected with a certain probability, and the selected gene is suddenly selected. Mutations may be made. For example, in FIGS. 20A to 20C, the value of the evaluation function J of the individual GS _n is set to 0 or less. As shown in FIG. 20A, when mutation is performed with respect to the rotation angle θ, the gene of the rotation angle θ is set to a random number within the range of −90 ° ≦ θ ≦ 90 ° for the individual GS _n . . As a result of the mutation, a new individual GS _{n ′} is generated. As shown in FIG. 20 (b), when a mutation is performed with respect to the first thickness coefficient c, the gene of the first thickness coefficient c is assigned to the individual GS _n within the range of 1 <c <xMax / β. As a result of mutation, a new individual GS _{n ′} is generated. As shown in FIG. 20C, when mutation is performed with respect to the rotation angle θ and the first thickness coefficient c, the gene of the rotation angle θ is set to −90 ° ≦ θ ≦ 90 ° with respect to the individual GS _n . A random number in the range is used, and the gene of the first thickness coefficient c is a random number in the range of 1 <c <xMax / β. As a result of the mutation, a new individual GS _{n ′} is generated.

Next, in step S108, the determination unit 68 evaluates fitness. In the fitness evaluation here, the determination unit 68 is the same as the fitness evaluation in step S104 for the n individuals selected in step S105 and the new individuals generated in steps S106 and S107. Then, the value of each evaluation function J is calculated. Then, the determination unit 68 performs sorting in the order from the smallest number of the second support 30S among individuals whose evaluation function J is greater than zero. That is, the determination unit 68 positions the superior individual as the superior individual as the number of the second support 30S is smaller among the individuals whose evaluation function J is greater than zero. The top individual is the best individual. Below, to the best individual after sorting and individual G3 _1.

Next, in step S109, the determination unit 68 determines the end of the genetic algorithm. Here, in step S108, determining unit 68, the most evaluation minute rotational angle θ a predetermined angle of higher was best individuals G3 _1, create a new modified individuals. This predetermined angle is stored in advance in the storage unit 52. The value of the predetermined angle is not particularly limited, and is, for example, + 2 ° or −2 °. For example, the rotation angle of the best individual G3 ₁ is shifted by −2 °, and the individual is set as an individual G3 _{1 ′} . Next, the number of second supports 30S of the individual G3 _{1 ′} is calculated. Thereafter, the determination unit 68 calculates the value of the evaluation function J for the individual G3 _{1 ′} . When the value of the evaluation function J of the individual G3 _{1 ′} is larger than 0, the determination unit 68 determines the number of the second support 30S of the best individual G3 ₁ in step S108 and the second support 30S of the individual G3 _{1 ′} . Compare the number.

When the number of second supports 30S of the individual G3 _{1 ′} is smaller than the number of second supports 30S of the individual G3 ₁ , the determination unit 68 sets the determination result to NO, returns to Step S105, and repeats the steps after Step S105 again. Do. At this time, in step S105, among the n individuals selected in the previous step S105, the new individuals generated in steps S106 and S107, and the individual G3 _{1 ′} generated in step S108. N individuals with high evaluation are selected.

On the other hand, _'if the value of the evaluation function J is 0 or less, or, individuals G3 _1' individuals G3 ₁ greater than the evaluation function value of J is zero, individuals G3 ₁ number individuals G3 ₁ of the second support 30S _' If greater than or equal to the number of the second support 30S, determining unit 68, the determination result is YES, and to a solution of best individuals G3 _1. Here, the orientation of the target shaped object model 72 that is rotated by the main shaft 78 to the shaft at a rotation angle θ identified in best individuals G3 ₁ is considered optimal orientation of the target shaped object 70. Then, a value obtained by multiplying the thickness β of the second support 30S in the first thickness coefficient c identified in best individuals G3 ₁ becomes the thickness of the first support 30L.

  In the present embodiment, according to the position and orientation of the target object 70 determined by the arrangement determining device 100, the arrangement position of the second support 30S, and the thickness of the first support 30L, the entire object 20 is placed below the holder 13. Is shaped. That is, the support 30 is modeled under the holder 13 and the target model 70 is modeled under the support 30.

  As described above, in the present embodiment, as shown in FIG. 15, the first area 91 of the arrangement surface 82 is an area near the center of gravity 74 of the arrangement surface 82. Therefore, by making the thickness of the first support 30L added and arranged in the first region 91 larger than the thickness of the second support 30S added and arranged in the second region 92, the support 30 The load of the target model 70 can be supported around the center of gravity 74 of the target model 70. Therefore, the support 30 can stably support the load of the target object 70. In the present embodiment, the support 30 can support the load of the target object 70, and the number of second supports 30S added and arranged in the second region 91 of the arrangement surface 82 is the minimum. Such a rotation angle θ is an angle to be rotated about the main axis 78 of the target object model 72 arranged at the reference position and orientation. The support 30 can support the load of the target model 70 by modeling the target model 70 in the direction determined as described above. In addition, since the number of the second supports 30S can be reduced, it is possible to reduce the material necessary for modeling the support 30.

  For example, when all possible combinations of the rotation angle θ and the first thickness coefficient c are obtained and the optimum orientation of the target object 70 and the optimum thickness of the first support 30L are obtained, the calculation process is performed. It becomes enormous and the calculation load increases. However, in the present embodiment, the combination of the rotation angle θ and the first thickness coefficient c is determined by applying a genetic algorithm. Therefore, by applying the genetic algorithm, it is possible to prevent the calculation time from becoming enormous as compared with the case of obtaining all possible combinations of the rotation angle θ and the first thickness coefficient c. As a result, the calculation load can be reduced.

  In the present embodiment, the determination unit 68 calculates the value of the evaluation function J in each individual using the evaluation function J as shown in the above equation (7). Then, the determination unit 68 uses an individual whose value of the evaluation function J is greater than 0 and the number of the second supports 30S as a solution, and determines the optimum rotation angle θ and the thickness of the first support 30L. To decide. In this way, by using the evaluation function J as shown in the above formula (7), the target is oriented in a direction that can withstand the larger load of the load Gt and the adhesion force As of the target model 70 to be modeled. A model 70 can be set. Further, the first support 30L can be added and arranged in the first region 91 with the optimal thickness of the first support 30L.

  The 3D modeling system 10 including the arrangement determining device 100 according to the first embodiment has been described above. The arrangement determining apparatus according to the present invention is not limited to the arrangement determining apparatus 100 according to the first embodiment, and can be implemented in various other forms. Next, another embodiment will be briefly described. In the following description, the same reference numerals are used for the same components as those already described, and the description thereof is omitted.

Second Embodiment
FIG. 21 is a perspective view showing the target modeled object 70e, and is a diagram showing a forbidden surface. As shown in FIG. 21, the target object 70e may have a decorative surface 71e on which decorations such as letters or patterns are applied. When modeling such a target modeled object 70e, it is preferable not to add and arrange the support 30 on the decorative surface 71e. This is because when the support 30 is peeled from the target model 70e, the decorative surface 71e of the target model 70e may be damaged. When modeling such an object model 70e, it is preferable that the support 30 is not added and arranged on the decorative surface 71e. Therefore, in the present embodiment, when the target modeled object 70e has a surface on which the support 30 should not be added and arranged (hereinafter referred to as a prohibited surface), the support 30 sufficiently loads the target modeled object 70e. In addition to being able to support, it is preferable to determine the position and orientation of the target object 70e so that the support 30 is not added and arranged on the prohibited surface. In FIG. 21, the decorative surface 71e is a prohibited surface.

  In this embodiment, even if it is the target modeling object 70e which has a prohibition surface, it applies to a surface of the target modeling model corresponding to the target modeling object 70e by applying a genetic algorithm in the same procedure as the said embodiment. It can be determined what support 30 is added and arranged. However, in the genetic algorithm, the range of the rotation angle θ that is an individual gene may be a range in which the support 30 is not formed on the prohibited surface 71e. In the present embodiment, for example, as shown in FIG. 21, in the target object model 72e set to the reference position and orientation, the forbidden surface 71e is set to the upper surface of the target object model 72. The arrangement surface 82e is the lower surface of the target model object 72e. Here, the range of the rotation angle θ is a range in which the forbidden surface 71e does not coincide with the arrangement surface 82. For example, the range of the rotation angle θ is −90 ° ≦ θ ≦ 90 °. Here, the rotation angle θ among the genes of each individual may be determined by a random number within the above range. Since other determination procedures are the same as those in the above embodiment, detailed description thereof is omitted.

  In the present embodiment, the rotation angle candidate generation unit 62 uses only the rotation angle θ as a candidate so as not to add and arrange the support 30 to the prohibited surface 71e where the support 30 is not added and arranged. Therefore, the rotation angle θ at which the support 30 is not formed on the prohibited surface 71e can be set to an angle that rotates from a predetermined reference direction. Therefore, the support 30 can support the load of the target object 70e without adding and arranging the support 30 on the prohibited surface 71e.

<Third Embodiment>
In each said embodiment, as shown to the said Formula (7), the thickness of the 1st support 30L was represented by the ratio with respect to the thickness of the 2nd support 30S. In other words, the thickness of the first support 30L is represented by the thickness β of the second support 30S × the first thickness coefficient c. Therefore, when applying the genetic algorithm, one of the genes of the individual is set as the first thickness coefficient c. However, one of the individual genes is not limited to the first thickness coefficient c.

  For example, when applying the genetic algorithm, the gene of the individual may be the rotation angle θ and the thickness of the first support 30L. Even in such a case, it is possible to determine the optimum rotation angle θ and the thickness of the first support 30L by applying the genetic algorithm in the procedure as shown in FIG.

<Modification>
In each of the embodiments described above, the main axis setting unit 58 sets the main axis 78 of the target object model 72 using the center of gravity of a plurality of polygons constituting the target object model 72 as shown in FIG. The main axis 78 is a line connecting the center of gravity 74 of the target object model 72 and the center of gravity of any of the plurality of polygons (here, the farthest point 76). However, the main axis 78 may be a line that connects the center of gravity 74 of the target object model 72 and the vertex farthest from the center of gravity 74 among the vertices of a plurality of polygons. In this case, the main axis setting unit 58 calculates the distance between the center of gravity 74 of the target model object 72 and the vertices of a plurality of polygons constituting the target model model 72. Then, the spindle setting unit 58 sets the vertex of the polygon corresponding to the longest distance among the plurality of distances between the center of gravity 74 of the target object model 72 and the vertices of the plurality of polygons as the farthest point. A line connecting the far point and the center of gravity 74 of the target object model 72 may be set as the main axis 78.

  In addition, you may combine each said embodiment and modification suitably.

  As described above, the storage unit 52, the preprocessing unit 54, the center of gravity calculation unit 56, the initial unit 57, the spindle setting unit 58, the reference unit 60, the rotation angle candidate generation unit 62, and the first thickness candidate generation of the arrangement determining apparatus 100. The unit 63, the arrangement plane determining unit 64, the reference point setting unit 65, the region dividing unit 66, the number calculating unit 67, and the determining unit 68 may be configured by software. In other words, each of the above units may be realized by a computer by reading the computer program into the computer. The present invention includes a computer program for causing a computer to function as each unit described above. The present invention also includes a computer-readable recording medium on which the computer program is recorded.

DESCRIPTION OF SYMBOLS 10 3D modeling system 10A 3D modeling apparatus 13 Holder 30 Support (support modeling object)
30L First Support 30S Second Support 52 Storage Unit 56 Center of Gravity Calculation Unit 58 Spindle Setting Unit 60 Reference Unit 62 Rotation Angle Candidate Generation Unit 63 First Thickness Candidate Generation Unit 64 Arrangement Surface Determination Unit 65 Reference Point Setting Unit 66 Region Division Unit 67 Number Calculation Unit 68 Determination Unit 70 Target Model 72 Target Model Model (3D Model)
74 Center of gravity 86 Reference point 91 First area 92 Second area 100 Arrangement determining device (support arrangement determining device)

Claims (9)

In a three-dimensional modeling apparatus that adds and arranges a support to a target modeling object that is a modeling target, and models the target modeling object and the support, the arrangement determining apparatus determines the arrangement position of the support,
The support includes a first support having a first thickness and a second support having a second thickness smaller than the first thickness.
A storage unit for storing data of a three-dimensional model of the target object;
A centroid calculating unit for calculating the centroid of the three-dimensional model;
A main axis that calculates a farthest point farthest from the center of gravity of the three-dimensional model among points on the outer peripheral surface of the three-dimensional model, and sets a main axis that connects the center of gravity of the three-dimensional model and the farthest point A setting section;
When the position and orientation of the three-dimensional model in which the main axis is parallel to a horizontal plane are set to a predetermined reference position and orientation, the predetermined reference position and orientation are stored in the storage unit. A reference unit for moving and rotating the three-dimensional model;
A rotation angle candidate generating unit that generates a plurality of rotation angle candidates for rotating the three-dimensional model about the main axis with reference to the three-dimensional model arranged so as to have the predetermined reference position and orientation; ,
In each of the three-dimensional models rotated based on the plurality of generated rotation angles, an arrangement surface determination unit that determines an arrangement surface on which the support is added and arranged;
A reference point setting unit that draws a perpendicular line from the center of gravity to the arrangement surface and sets a reference point at which the perpendicular and the arrangement surface intersect;
In each of the arrangement planes, a first area in which a predetermined number of the first supports are added and arranged within a predetermined range including the reference point set by the reference point setting unit; An area dividing unit that divides the area excluding the first area into a second area where the second support is added and arranged;
A first thickness candidate generator that creates a plurality of candidates for the first thickness that is the thickness of the first support;
A number calculation unit that calculates the number of the second supports when the second support is added to and arranged in the second region at intervals in each of the arrangement surfaces determined by the arrangement surface determination unit; ,
Among the plurality of rotation angles generated by the rotation angle candidate generation unit and the first thicknesses of the plurality of first supports generated by the first thickness candidate generation unit, the support is the A determination unit that is capable of supporting a load of a target object and that determines the rotation angle that minimizes the number of the second supports and the first thickness of the first support; ,
An apparatus for determining the placement of a support.   The determination unit generates an individual having the rotation angle generated by the rotation angle candidate generation unit and the first thickness generated by the first thickness candidate generation unit as a gene, and a genetic algorithm And determining the rotation angle and the first thickness such that the support can support the load of the target object and the number of the second supports is minimized. The support arrangement determining apparatus according to claim 1. The evaluation function J used in the genetic algorithm is such that the number of the first support is p, the thickness of the first support is x, the number of the second support is q, the thickness of the second support is β, When the tensile strength of one of the supports is k, the load of the target object is Gt, the adhesion between the target object and the three-dimensional modeling apparatus is As, and the safety factor is S,
J = (x.k.p + .beta..k.q) / S-max (Gt, As)
Represented by
The determination unit combines the plurality of rotation angles generated by the rotation angle candidate generation unit and the first thicknesses of the plurality of first supports generated by the first thickness candidate generation unit. A plurality of individuals are generated, and values obtained from the evaluation function are calculated for each of the plurality of generated individuals, so that the value of the evaluation function is greater than 0 and the number of the second supports is minimized. The support arrangement determination device according to claim 2, wherein the individual is extracted, and the rotation angle and the first thickness of the first support are determined. The target object has a forbidden surface that does not add and place the support;
4. The support arrangement determination device according to claim 1, wherein the rotation angle candidate generation unit does not use the rotation angle as a candidate for adding and arranging the support on the prohibited surface. 5. The three-dimensional modeling apparatus;
A support arrangement determining device according to any one of claims 1 to 4,
3D modeling system. In a three-dimensional modeling apparatus that adds and arranges a support to a target modeling object that is a modeling target, and models the target modeling object and the support, it is an arrangement determination method that determines the arrangement position of the support,
The support includes a first support having a first thickness and a second support having a second thickness smaller than the first thickness.
A centroid calculating step of calculating the centroid of the three-dimensional model;
A main axis that calculates a farthest point farthest from the center of gravity of the three-dimensional model among points on the outer peripheral surface of the three-dimensional model, and sets a main axis that connects the center of gravity of the three-dimensional model and the farthest point A setting process;
When the position and orientation of the three-dimensional model in which the main axis is parallel to the horizontal plane are set to a predetermined reference position and orientation , the three-dimensional model is moved and adjusted to the predetermined reference position and orientation. A reference process to rotate;
A rotation angle candidate generating step for generating a plurality of rotation angle candidates for rotating the three-dimensional model around the main axis with reference to the three-dimensional model arranged so as to have the predetermined reference position and orientation; ,
In each of the three-dimensional models rotated based on the plurality of generated rotation angles, an arrangement surface determination step for determining an arrangement surface on which the support is added and arranged;
A reference point setting step of drawing a perpendicular line from the center of gravity to the arrangement surface and setting a reference point where the perpendicular line and the arrangement surface intersect;
In each of the arrangement planes, a first area in which a predetermined number of the first supports are added and arranged in an area within a predetermined range including the reference point set in the reference point setting step; A region dividing step of dividing the region into a second region where the second support is added and arranged, excluding the first region;
A first thickness candidate generation step of creating a plurality of candidates for the first thickness that is the thickness of the first support;
In each of the arrangement planes determined in the arrangement plane determination step, a number calculation step of calculating the number of the second supports when the second support is added and arranged in the second region with an interval between them,
Among the plurality of rotation angles generated in the rotation angle candidate generation step and the first thicknesses of the plurality of first supports generated in the first thickness candidate generation step, the support is the target modeling. A determination step of determining the rotation angle that can support a load of an object and that minimizes the number of the second supports, and the first thickness of the first support;
A method for determining the placement of a support.   In the determining step, an individual is generated using the rotation angle generated in the rotation angle candidate generation step and the first thickness generated in the first thickness candidate generation step as a gene, and a genetic algorithm is used. The rotation angle and the first thickness are determined so that the support can support the load of the target object and the number of the second supports is minimized. 6. A support arrangement determining method described in 6. The evaluation function J used in the genetic algorithm is such that the number of the first support is p, the thickness of the first support is x, the number of the second support is q, the thickness of the second support is β, When the tensile strength of one of the supports is k, the load of the target object is Gt, the adhesion between the target object and the three-dimensional modeling apparatus is As, and the safety factor is S,
J = (x.k.p + .beta..k.q) / S-max (Gt, As)
Represented by
In the determination step, a plurality of the rotation angles generated in the rotation angle candidate generation step and a plurality of the first thicknesses of the first support generated in the first thickness candidate generation step Generating an individual, calculating a value obtained from each of the generated evaluation functions of the plurality of individuals, wherein the value of the evaluation function is greater than 0, and the number of the second supports is minimized The support arrangement determining method according to claim 7, wherein an individual is extracted, and the rotation angle and the first thickness of the first support are determined. The target object has a forbidden surface that does not add and place the support;
The support arrangement determination method according to any one of claims 6 to 8, wherein, in the rotation angle candidate generation step, the rotation angle that adds and arranges the support on the prohibited surface is not used as a candidate.

Images