JP6939487B2 - Structure model generator, particle behavior analysis device, structure model generation method, particle behavior analysis method, and program - Google Patents

Description

The present invention relates to a structure model generation device, a particle behavior analysis device, a structure model generation method, a particle behavior analysis method, and a program, and in particular, the behavior of particles using DEM (Discrete Element Method). It is suitable for use in analyzing.

There is DEM as a method for analyzing the behavior of particles flowing on the surface of various structures. In DEM, it is necessary to model the structure. Patent Document 1 discloses a method of expressing the surface of a structure by arranging the wall particles inside the surface of the structure so as to be in contact with the surface of the structure.

Japanese Unexamined Patent Publication No. 2012-37937 Japanese Unexamined Patent Publication No. 2011-69656

Fausto Bernardini et al. The Ball Pivoting Algorithm for Surface Reconstruction. IEEE Transactions on Visualization and Computer Graphics 5, 4 (Oct.-Dec. 1999), 349-359. Yan DM, Guo JW, Wang B et al. A survey of blue-noise sampling and its applications. JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY 30 (3): 439-452 Massimiliano Corsini et al. Efficient and Flexible Sampling with Blue Noise Properties of Triangular Meshes. IEEE Transactions on Visualization and Computer Graphics, 18, (2012) 914-924 Brian Curless and Marc Levoy. A volumetric method for building complex models from range images. In Proceedings of SIGGRAPH '96, August 1996, 303-312

However, in the technique described in Patent Document 1, the coordinates of the points on the surface of the structure are input, the area that can be approximated to a straight line on the surface of the structure is specified from the input coordinates, and the specified area is used for the wall. Place the particles. Therefore, it is necessary to generate information on the coordinates of points on the surface of the structure, convert the coordinates into information on the surface of the structure, and arrange wall particles. The number of coordinates of points on the surface of a structure is hundreds of thousands to tens of millions, and in some cases even more. In the technique described in Patent Document 1, it is necessary to convert a large number of these point clouds into information on the surface of the structure, which requires a great deal of calculation cost. Further, the coordinates of the points on the surface of the structure can be obtained by the three-dimensional measurement, but the result of the three-dimensional measurement (coordinates of the points on the surface of the structure) may include measurement noise. In such a case, in the technique described in Patent Document 1, even if the coordinates of the points on the surface of the structure are the coordinates of the positions far from the actual structure, the coordinates of the surface of the structure are taken into consideration. May generate information.

The present invention has been made in view of the above problems, and when analyzing the behavior of particles using DEM, the calculation load on the wall particles representing the structure with which the particles are in contact is increased. The purpose is to arrange without doing.

The structure model generator of the present invention uses a plurality of wall particles as a model of the structure when analyzing the behavior of a plurality of analysis particles in the region where the structure exists by the discrete element method. It is a structure model generation device that is generated by The point group data acquisition means to be performed and a part of a plurality of points included in the point group data are set as analysis target points, and data including the positions of the plurality of analysis target points is extracted as analysis target point group data. Wall particle setting for setting wall particles for each of the plurality of analysis target points included in the analysis target point group data using the point group data extraction means to be performed and the analysis target point group data. The point group data extraction means is adjacent to each other at the plurality of analysis target points, rather than a variation in the distance between the points adjacent to each other at the plurality of points included in the point group data. It is characterized in that data including the positions of the plurality of analysis target points is extracted as analysis target point group data so that the variation of the point distances becomes small.

The particle behavior analysis device of the present invention includes an analysis particle behavior analysis means for analyzing the behavior of a plurality of analysis particles in a region where a structure exists by the discrete element method, and the structure model generation device. The particle behavior analysis device has the particle behavior analysis means for analysis, and the analysis particles come into contact with the wall particles set by the wall particle setting means included in the structure model generation device. When it is determined, it is characterized in that the force received from the wall particles to the analysis particles is applied to the analysis particles to move the analysis particles.

The structure model generation method of the present invention uses a plurality of wall particles as a model of the structure when analyzing the behavior of a plurality of analysis particles in the region where the structure exists by the discrete element method. The structure model generation method for generating the structure, wherein the point group data acquisition means is data measured by a three-dimensional measuring device and includes the positions of a plurality of points indicating the positions on the surface of the structure. The point group data acquisition process for acquiring data as point group data and the point group data extraction means set some points among the plurality of points included in the point group data as analysis target points, and a plurality of the analysis targets. A point group data extraction step for extracting data including the position of a point as analysis target point group data, and a plurality of wall particle setting means included in the analysis target point group data using the analysis target point group data. Each of the analysis target points of the above has a wall particle setting step for setting the wall particles, and the point group data extraction steps are mutually performed at the plurality of points included in the point group data. Information including the positions of the plurality of analysis target points is used as analysis target point group data so that the variation in the distances of the plurality of analysis target points at the plurality of analysis target points becomes smaller than the variation in the distances of the adjacent points. It is characterized by extracting.

Particle behavior analysis method of the present invention, the analysis particle behavior analyzer, and analyzing the particle behavior analysis step of analyzing the behavior of a plurality of analysis particles in the region where the structure is present by the discrete element method, the It is a particle behavior analysis method having each step of the structure model generation method, and in the analysis particle behavior analysis step, the analysis particles are set in the wall particle setting step in the structure model generation method. When it is determined that the particles are in contact with the wall particles, the force received by the analysis particles from the wall particles is applied to the analysis particles to move the analysis particles.

The first example of the program of the present invention is characterized in that a computer functions as each means of the structure model generator.
A second example of the program of the present invention is characterized in that a computer functions as each means of the particle behavior analysis apparatus.

According to the present invention, when analyzing the behavior of particles using a DEM, wall particles representing a structure in contact with the particles can be arranged without increasing the calculation load.

FIG. 1 is a diagram showing an example of a functional configuration of a particle behavior analysis device. FIG. 2 is a diagram showing an example of a schematic configuration of a turning chute handled by the point cloud data acquisition unit. FIG. 3 is a diagram showing an example of arrangement of point cloud data acquired by the point cloud data acquisition unit. FIG. 4 is a diagram showing an example of arrangement of point cloud data not removed by the point data removal unit among the point cloud data acquired by the point cloud data acquisition unit. FIG. 5 is a diagram conceptually showing an example of the arrangement of analysis target points. FIG. 6 is a diagram conceptually showing an example of a method of setting wall particles. FIG. 7 is a diagram showing an example of arrangement of wall particles. FIG. 8 is a diagram conceptually showing an example of the appearance of two particles (analytical particles and wall particles) in contact with each other and an example of the behavior of the analysis particles. FIG. 9 is a diagram showing an example of the behavior of particles analyzed by DEM. FIG. 10 is a flowchart illustrating an example of processing by the particle behavior analysis device when setting wall particles. FIG. 11 is a flowchart illustrating an example of processing of the particle behavior analysis apparatus when analyzing the behavior of the particles for analysis. FIG. 12 is a diagram showing another example of arrangement of analysis target points.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a diagram showing an example of a functional configuration of the particle behavior analysis device 100.

The particle behavior analysis device 100 can be realized by using, for example, an information processing device (personal computer) provided with a CPU, ROM, RAM, HDD, various interfaces, a display, or the like, or dedicated hardware. The functions of the particle behavior analysis device 100 will be described in detail below. In the present embodiment, a case where a swivel chute for a bellless type furnace top charging device of a blast furnace is expressed as a structure will be described as an example. Therefore, the particles whose behavior is to be analyzed (hereinafter, these particles are referred to as analysis particles as necessary) are sinter or coke. However, it goes without saying that the structure and the particles for analysis are not limited to such a thing.

(Point cloud data acquisition unit 101)
The point cloud data acquisition unit 101 acquires point cloud data, which is a set of data indicating the coordinates of points on the surface of the wall (for example, the coordinates of a three-dimensional Cartesian coordinate system). The point cloud data can be measured, for example, by performing a known three-dimensional measurement. For example, point cloud data is measured by a method obtained from the measurement result of a three-dimensional laser scanner for a swirling chute, a method obtained from a measurement result of a stereo camera for a swirling chute, or a method of applying a pattern projection method to a swirling chute. can do. In the measurement by the three-dimensional laser scanner, the object is irradiated with the laser beam in a radial pattern, and the object is based on the time from the irradiation of the laser beam to the reflection and reception of the object and the irradiation angle of the laser beam. 3D measurement (measurement of coordinates of points on a 3D surface). In the measurement with a stereo camera, three-dimensional measurement (measurement of the coordinates of points on the three-dimensional surface) of the object is performed by the principle of triangulation based on images obtained by capturing the same position of the object from different viewpoints. .. In the pattern projection method, an image of an object is imaged with the pattern light projected onto the object, and the three-dimensional coordinates of the points on the pattern reflected in the image are obtained by the principle of triangulation. Since any of the methods can be realized by a known technique, detailed description thereof will be omitted here.

For example, when the particle behavior analysis device 100 and the three-dimensional measurement device that performs three-dimensional measurement are communicably connected to each other, the point cloud data acquisition unit 101 receives the point cloud data transmitted from the three-dimensional measurement device. The point cloud data can be acquired by storing the data in the storage medium. Further, the point cloud data acquisition unit 101 acquires the point cloud data by reading the point cloud data from the portable storage medium that stores the point cloud data measured by the three-dimensional measuring device and storing the point cloud data in the storage medium. Can be done. The number of point cloud data can range from hundreds of thousands to tens of millions or more.

FIG. 2 is a diagram showing an example of a schematic configuration of a turning chute handled by the point cloud data acquisition unit 101. FIG. 2A is a diagram showing an example of a swivel chute region for acquiring point cloud data by the point cloud data acquisition unit 101. FIG. 3 is a diagram showing an example of arrangement of point cloud data acquired by the point cloud data acquisition unit 101. In FIG. 3, although it is not clear due to the convenience of notation, the parts shown in gray and black in FIG. 3 are a set of points.

(Point data removal unit 102)
The point data removal unit 102 determines whether or not there is a point to be removed in the point cloud data acquired by the point cloud data acquisition unit 101. The point to be removed is, for example, a point in a region of the structure (a swirling chute in the present embodiment) that does not contribute to the flow of particles (that is, a region that does not correspond to the path of the particles). The point data removing unit 102 may automatically determine whether or not there is a point to be removed according to a predetermined condition, and whether or not there is a point to be removed based on an input operation to the particle behavior analysis device 100 by the operator. It may be determined whether or not.

When it is automatically determined whether or not there is a point to be removed according to a predetermined condition, for example, the point data removing unit 102 uses the point cloud data acquired by the point cloud data acquisition unit 101 to obtain a structure (this). In the embodiment, the outer shape of the swivel chute) is specified, and when it is determined that there is an area corresponding to a predetermined portion in the area within the specified outer shape, it is determined that there is a point to be removed, and when not, it is determined. , It can be determined that there is no point to remove.

When determining whether or not there is a point to be removed based on an input operation to the particle behavior analysis device 100 by the operator, for example, the point data removal unit 102 is information on the point cloud data acquired by the point cloud data acquisition unit 101. Is displayed on the computer display. When the operator determines from this display that there is a point to be removed in the point cloud data, the operator inputs the information of the point to be removed by operating the user interface of the particle behavior analysis device 100. When this information is input, the point data removing unit 102 can determine that there is a point to be removed, and if not, it can determine that there is no point to be removed.

When the point data removing unit 102 determines that there is a point to be removed in the point cloud data acquired by the point cloud data acquisition unit 101, the point data removing unit 102 identifies the point to be removed and extracts the data of the specified point from the point cloud data. Remove. The point data removing unit 102 may automatically specify the points to be removed according to predetermined conditions, or may specify the points to be removed based on the input operation to the particle behavior analysis device 100 by the operator.

When the points to be removed are automatically specified according to a predetermined condition, for example, the point data removing unit 102 is included in the region within the outer shape of the structure (swivel chute in the present embodiment) specified as described above. , A point in the region of a predetermined portion can be specified as a point to be removed. When specifying the points to be removed based on the input operation to the particle behavior analysis device 100 by the operator, the point data removing unit 102 removes from the information of the points to be removed input by the operation of the user interface by the operator as described above. It is possible to specify the point to be done.

FIG. 2B is a diagram showing an example of a region in which a point to be removed exists in the region of the swivel chute. In FIG. 2B, points existing in the shaded area are specified as points to be removed. FIG. 2C is a diagram showing an example of a swivel chute region used for analysis. In FIG. 2B, the attachment portion of the swirl chute and the proximal end portion are portions that do not contribute to the flow of the raw material (particles), and therefore this portion is not used in the analysis. Therefore, as shown in FIG. 2C, the region of the swivel chute excluding this portion is defined as the swirl chute region used for the analysis. FIG. 4 is a diagram showing an example of arrangement of point cloud data not removed by the point data removal unit 102 among the point cloud data acquired by the point cloud data acquisition unit 101. In FIG. 4, although it is not clear due to the convenience of notation, the parts shown in gray and black in FIG. 4 are a set of points. In the following description, among the point cloud data acquired by the point cloud data acquisition unit 101, the point cloud data not removed by the point data removal unit 102 will be referred to as analysis target candidate point cloud data, if necessary. Further, the points not removed by the point data removing unit 102 are referred to as analysis target candidate points as necessary.

(Point data complement unit 103)
Depending on the three-dimensional measurement method, there may be a region that becomes a blind spot, so there may be a region where the density of the analysis target candidate points (the number of points per unit area) included in the analysis target candidate point cloud data is low. be. Therefore, the point data complementing unit 103 determines whether or not there is a region for complementing points in the region where the analysis target candidate point cloud data exists. The area in which the analysis target candidate point cloud data exists is an area including all the analysis target candidate point cloud data. For example, the area surrounded by connecting the outermost analysis target candidate points so as to include all the analysis target candidate point cloud data can be set as the area in which the analysis target candidate point cloud data exists. The point data complementing unit 103 may automatically determine whether or not there is a region for complementing points according to predetermined conditions, and complements points based on an input operation to the particle behavior analysis device 100 by the operator. It may be determined whether or not there is an area.

When it is automatically determined whether or not there is an area for complementing points according to a predetermined condition, for example, the point data complementing unit 103 has a point density of the area in which the analysis target candidate point cloud data exists. If there is an area equal to or less than a predetermined value, it can be determined that there is an area that complements the points, and if not, it can be determined that there is no area that complements the points.

When determining whether or not there is a region for complementing points based on an input operation to the particle behavior analysis device 100 by the operator, for example, the point data complementing unit 103 displays information on the analysis target candidate point cloud data on the computer display. do. When the operator determines from this display that there is a point-complementing area in the area where the analysis target candidate point cloud data exists, the operator operates the user interface of the particle behavior analysis device 100 with the information of the point-complementing area. And enter. When this information is input, the point data complementing unit 103 can determine that there is an area for complementing points, and if not, it can determine that there is no area for complementing points.

When the point data complementing unit 103 determines that there is an area for complementing points in the area where the analysis target candidate point cloud data exists, the point data is added to the area, and the point data is used as the analysis target candidate. Add to the point.
In the present embodiment, the point data complementing unit 103 creates a mesh based on the point data included in the area that complements the points, and adds the points inside the mesh. As a method for producing the mesh, for example, the Ball Pivoting method described in Non-Patent Document 1 can be used. The Ball Pivoting method is a triangle in which a ball (sphere) with a predetermined radius is rotated and moved in a region where a point group exists, three points that touch the ball are selected without including other points, and the selected three points are the vertices. It is a method of repeatedly generating the mesh of. Since the Ball Pivoting method is a known technique described in Non-Patent Document 1 and the like, detailed description thereof will be omitted here. Further, the method for creating the mesh is not limited to the Ball Pivoting method, and may be another known method (for example, the VCG method described in Non-Patent Document 4).

Further, in the present embodiment, the point data complementing unit 103 has a blue-noise property so that the spatial frequency distribution of the points after adding the points to the region (mesh) that complements the points has a blue-noise property. Add point data to the area that complements the points. The spatial frequency of points (dots) is obtained by Fourier transforming the coordinates of the points, the global spacing of the point cloud is represented as a low frequency component, and the difference in detail spacing is represented as a high frequency component. The blue noise characteristic refers to a characteristic in which the spatial frequency distribution of points (dots) has more high-frequency components than low-frequency components and does not have a maximum value. As described in Non-Patent Document 2, the pattern of arrangement of points having a blue noise characteristic is randomized uniform distributions, and is excellent in the dispersibility of points. As described above, the distribution of the spatial frequency of the point shows the relationship between the spatial frequency and the spectral intensity obtained by Fourier transforming the coordinates of the point. In general, the spatial frequency of a point (dot) corresponds to the reciprocal of the distance between two adjacent points, and the spatial frequency distribution of the point (dot) is that of each spatial frequency (the distance). Corresponds to the frequency of appearance.

For example, there is a Poisson disk subsampling method as a method of arranging points so that the distribution of spatial frequencies has a blue noise characteristic. The Poisson disk subsampling method selects the position of a point according to a random number, and repeats placing the selected point when the distance between the position of the selected point and the position of the already placed point is more than a predetermined distance. This is the method to be performed. The distribution of points arranged by the Poisson disk subsampling method is a random distribution with the additional constraint that the distance between adjacent points must not be less than a predetermined distance. Since the Poisson disk subsampling method is a known technique described in Non-Patent Documents 2, 3 and the like, detailed description thereof will be omitted here. The method of arranging the points so that the spatial frequency distribution has the blue noise characteristic is not limited to the Poisson disk subsampling method, and other known methods (for example, the Relaxation-Based Sampling method described in Non-Patent Document 2). , Patch / Tile-Based Sampling method, etc.).

(Point cloud data extraction unit 104)
The point cloud data extraction unit 104 extracts points for arranging wall particles, which will be described later, from the analysis target candidate point cloud data. In the following description, this point is referred to as an analysis target point as necessary, and the point cloud data which is a set of these points is referred to as an analysis target point cloud data as necessary. In the present embodiment, the point cloud data extraction unit 104 extracts the analysis target point from the analysis target candidate point cloud data so that the spatial frequency distribution of the analysis target point has a blue noise characteristic. FIG. 5 is a diagram conceptually showing an example of the arrangement of analysis target points. At this time, the point cloud data extraction unit 104 may extract the analysis target points so that the density of the analysis target points is within a predetermined range. As will be described later, in the present embodiment, the surface (wall) of the structure (swivel chute in the present embodiment) is expressed by arranging one spherical particle for one analysis target point. In the following description, the spherical particles will be referred to as wall particles, if necessary. If the density of the analysis target points is too high, a large number of wall particles will be arranged, which will increase the calculation load. On the other hand, if the density of the analysis target points is too low, the surface of the structure (swivel chute in the present embodiment) may not be accurately reproduced. The predetermined range described above is determined from such a viewpoint.

As described above, in the present embodiment, the point data complementing unit 103 complements the points so that the spatial frequency distribution of the points after adding the points to the region (mesh) that complements the points has a blue noise characteristic. Add point data to the area. Therefore, the point cloud data extraction unit 104 may adopt the analysis target candidate point cloud data in the region that complements the points as the analysis target point cloud data as it is.

Further, the point data complementing unit 103 adds point data to the area that complements the points so that the spatial frequency distribution of the points after the addition to the area (mesh) that complements the points does not have blue noise characteristics. You may. For example, the point data complementing unit 103 may randomly arrange the point data in the area that complements the points. In this case, the point cloud data extraction unit 104 determines that the spatial frequency distribution of the analysis target points is blue noise from all the analysis target candidate point cloud data including the analysis target candidate point cloud data in the region that complements the points. The analysis target point may be extracted so as to have characteristics. The reason why it is preferable that the spatial frequency distribution of the analysis target points has a blue noise characteristic will be described later.

(Surface roughness information acquisition unit 105)
The surface roughness information acquisition unit 105 sets the maximum value of the amount of dent from the surface of the structure (swivel chute in this embodiment) (maximum surface roughness value ζ _1,2 [m] in FIG. 6) of the structure. It is acquired as surface roughness information indicating the surface roughness of. The surface roughness information acquisition unit 105 may acquire the surface roughness information based on an input operation to the particle behavior analysis device 100 by the operator, or may automatically set the surface roughness information. When the surface roughness information is automatically set, for example, the surface roughness information acquisition unit 105 includes physical quantities reflecting the state of surface wear of the structure (swivel chute in this embodiment) and surface roughness information. Are created in association with each other, and when the physical quantity obtained from the actual measurement and the information indicating the position of the structure (swivel chute in the present embodiment) are input, the table is associated with the physical quantity. The surface roughness information stored in can be set as the surface roughness information at the position. Further, in the present embodiment, the surface roughness information acquisition unit 105 acquires the surface roughness information for each region where the frictional resistance can be approximated to be the same.

(Wall particle setting unit 106)
The wall particle setting unit 106 sets 1 wall particle for each of the analysis target points extracted by the point cloud data extraction unit 104, based on the surface roughness information acquired by the surface roughness information acquisition unit 105. Set one by one. In the present embodiment, the wall particle setting unit 106 sets spherical particles centered on the position of the analysis target point as wall particles. FIG. 6 is a diagram conceptually showing an example of a method of setting wall particles.

In FIG. 6, first, the wall particle setting unit 106 derives the distance P between the analysis target point 601 and the analysis target point 602 adjacent to the analysis target point 601. The wall particle setting unit 106 derives such a distance P for all analysis target points adjacent to the analysis target point 601. Next, the wall particle setting unit 106 acquires _{surface roughness information (maximum surface roughness value ζ i, j} ) corresponding to the positions where the analysis target points 601 and 602 exist. _{The wall particle setting unit 106 provides surface roughness information (maximum surface roughness value ζ i,)} corresponding to each position where the analysis target point 601 and all the analysis target points adjacent to the analysis target point 601 exist. _j ) is obtained.

Then, the wall particle setting unit 106 sets the radius of the sphere centered on the analysis target point 601 to r _wi [m], and sets the radius of the sphere centered on the analysis target point 602 adjacent to the analysis target point 601 to r _{wj. Surface roughness information (maximum value of surface roughness ζ i)} corresponding to the distance P of the analysis target points 601 and 602 and the position where the analysis target points 601 and 602 exist in the following equation (1) set to [m]. _{, j} ) and substitute. In Eq. (1), i is a variable that specifies one of the two analysis target points adjacent to each other, and j is a variable that specifies the other analysis target point. Further, ζ _{i and j are surface roughness information (maximum surface roughness value ζ i} ) corresponding to the position where two analysis target points (analysis target points specified by variables i and j) that are adjacent to each other exist. _{, J} ). In the example shown in FIG. 6, i is a variable (= 1) that specifies the analysis target point 601 and j is a variable (= 2) that specifies the analysis target point 602.

The particle setting unit 106 for the wall performs the above substitution to the equation (1) for all the sets of two analysis target points adjacent to each other. As a result, Eq. (1) is obtained for the number of pairs of two analysis target points adjacent to each other. _{The wall particle setting unit 106 sets the radii r wi} and r _wj of the spheres satisfying the equation (1) thus obtained in order from, for example, the analysis target point (analysis target point 601) of i = 1. Calculate for the analysis target point.
Incidentally, in view of a later collision detection process and the calculation load of the mobile processing of each analysis particle radius r _wi of the sphere, preferably as the difference between the r _wj is small, the radius r _wi of these spheres, r _wj is equal to Is the most preferable.

If the sphere radii r _wi and r _wj are not uniquely determined as a result of this calculation, the wall particle setting unit 106 is specified, for example, by two analysis target points (variables i and j) adjacent to each other. _{By changing the value of the surface roughness information (maximum surface roughness value ζ i, j} ) corresponding to the position where the analysis target point) exists, for all the analysis target points, the sphere centered on the analysis target point The radii r _wi and r _{w j} can be uniquely determined. At this time, the wall particle setting unit 106 makes the _{amount of change with respect to the maximum value ζ i, j of the surface roughness and the number of the maximum value ζ i, j} of the surface roughness to be changed as small as possible. It is preferable to do so.

Further, in order to reduce the calculation load and simplify the operation of the wall particle setting unit 106, the same two-point distance P is used for all adjacent analysis target points, and the same sphere is used for all analysis target points. The radius r _{w of} may be used. Specifically, all pairs of two analysis target points adjacent to each other are extracted from all analysis target points, and all of the extracted two analysis target point pairs are two points of the two analysis target points. The distance may be obtained, and the average value of the distance between the two points to be analyzed may be given as the distance P, and the condition that _{the radius of the sphere r wi} = r _wj = r _{w is added in Eq. (1).} do it. When the distance P is obtained in this way, a representative value (median value or the like) other than the average value may be used. The distance P as the average value of the distances between the two points to be analyzed is a value for convenience for simple calculation, and it is not necessary to change the position of the analysis target points. When performing such a simple calculation in the wall particle setting unit 106, the surface roughness information (maximum surface roughness value ζ _{i, j} ) is not obtained for each pair of two analysis target points adjacent to each other. , Ζ may be used as a representative value of the amount of dent from the surface of the structure.

The wall particle setting unit 106 sets a sphere centered on the analysis target point and having a radius r _wi or r _wj calculated with respect to the analysis target point as wall particles for the analysis target point. The wall particle setting unit 106 sets such wall particles for each of all analysis target points. Specifically, the wall particle setting unit 106 sets information including the coordinates of the center and the radius as the information of the wall particles.

In the present embodiment, as shown in FIG. 6, the surface roughness information (maximum surface roughness value ζ _{1, 2} ) corresponding to the positions where two analysis target points 601 and 602 adjacent to each other exist is provided. , The distance between the point 605a and the point 606. The point 606 is a point on the surface side of the structure (swivel chute in the present embodiment) among the points where the wall particles 603 and 604 intersect. The point 605a is a point on the virtual line 605 connecting the points 603a and 604a located on the surface of the structure (swivel chute in this embodiment) of the wall particles 603 and 604 with respect to the analysis target points 601 and 602. , An intersection with a perpendicular line extending from the point 606 in a direction perpendicular to the virtual line 605. In this way, the maximum value of the amount of dent in the gap formed between the wall particles 603 and 604 arranged adjacent to each other is the surface roughness information (surface roughness information) corresponding to the position where the analysis target points 601 and 602 exist. The value corresponds to the maximum roughness value ζ _1,2 ) (here, the value equal to the maximum surface roughness value ζ _1,2 ).
FIG. 7 is a diagram showing an example of arrangement of wall particles. In FIG. 7, although it is not clear for the convenience of notation, spherical wall particles as shown enlarged at the tip of the arrow line in FIG. 7 are arranged in the portion shown in black in FIG. 7.

(Attribute information setting unit 107)
The attribute information setting unit 107 inputs the attribute information of the structure (in this embodiment, the swirling chute) to the particle behavior analysis device 100 by the operator for each of the wall particles set by the wall particle setting unit 106. Get based on the operation. The attribute information set by the attribute information setting unit 107 is, for example, Young's modulus, Poisson's ratio, friction coefficient, and velocity. Here, the speed is set when the structure (swivel chute in this embodiment) moves.
Even if the attribute information setting unit 107 sets the attribute information individually for each of the wall particles, it can be approximated as having the same physical properties (for example, if the frictional resistance is the same). Attribute information may be set collectively for a plurality of wall particles belonging to (approximate region).

(Particle Behavior Analysis Unit 108 for Analysis)
The analysis particle behavior analysis unit 108 analyzes the behavior of spherical analysis particles by DEM. As described above, in the present embodiment, the analysis particles are sinter or coke. The analysis particle behavior analysis unit 108 arranges each analysis particle at a desired position based on the information of the analysis particles set based on, for example, an input operation to the particle behavior analysis device 100 by the operator. Then, the analysis particle behavior analysis unit 108 determines whether or not each analysis particle is in contact with other analysis particles or wall particles.
When the particle for analysis is in contact with another particle for analysis, the particle behavior analysis unit 108 for analysis finds the repulsive force in the normal direction and the shear direction, solves the translational equation of motion, and the force around the center of gravity. The equation of motion of rotation is solved by finding the moment of, and each of the particles for analysis is performed, and each particle for analysis is moved based on the result. On the other hand, when the particle for analysis is in contact with the particle for wall, the particle behavior analysis unit 108 for analysis finds the repulsive force in the normal direction and the shear direction, solves the translational equation of motion, and forces around the center of gravity. Solving the equation of motion of rotation by finding the moment of, is performed only for the particles for analysis, not for the particles for walls. For wall particles, these equations of motion may be solved, but for wall particles, the result (that is, force) of solving the equation of motion is prevented from acting, and the particles are moved by the force received from the analysis particles. Try not to.
In the present embodiment, the wall particles are prevented from moving due to the force received from the analysis particles, but it is not always necessary to do so. For example, the wall particles may vibrate with a minute amplitude due to the force from the analysis particles.

The analysis particle behavior analysis unit 108 performs such contact determination processing and movement processing of each analysis particle at predetermined time intervals. Then, from the result, the behavior of each analysis particle at the analysis time is obtained. It should be noted that the analysis particles move by giving a condition for disturbing the balance of the forces generated between the analysis particles during this analysis time.

_{Specifically, the analysis particle behavior analysis unit 108 obtains the repulsive forces F n and sw} in the normal direction in the contact region of the analysis particles s and w by the following equation (2), and obtains the following (3). _{), The repulsive force F t, sw} in the shear direction in the contact region of the particles s and w for analysis is obtained, the translational equation of motion of the following equation (4) is solved, and the rotational motion is obtained by the following equation (5). Solve the equation.

FIG. 8 shows an example of the appearance of the two particles (analytical particles and wall particles) 801 and 802 in contact with each other (FIG. 8 (a)) and an example of the behavior of the analytical particles 803 (FIG. 8 (FIG. 8). b)) is a diagram conceptually showing. The parameters of equations (2) to (5) will be described with reference to FIG.
In the equation (2), K _n is a spring constant in the normal direction, and can be set according to _{, for example, the distance H and δ n in FIG. 8 (a).} However, K _n can also be a preset constant value. Δun _{and sw} are relative translational displacements of the two particles s and w in the normal direction of the center of gravity. η _n is a viscosity constant in the normal direction and is preset. n _sw is a unit vector in the normal direction. Δt is a predetermined time interval and is set in advance.

In equation (3), min [A, B] means that the smaller of A and B is adopted. μ is a coefficient of friction and can be a preset value. However, μ can be different depending on the state and the like. t _sw is a unit vector in the shear direction. K _t is a spring constant in the shear direction, and can be set according to _{, for example, the distance H, δ n} , and δ _{t in FIG. 8 (a).} However, K _t can also be a preset constant value. Δut _{and sw} are relative translational displacements of the center of gravity between the two particles s and w in the shear direction. Δφ _sw is the relative rotational displacement of the contact region due to the rotation of the particles, and is represented by, for example, the difference between _{the rotational displacement φ s} of the particles 801 and the rotational displacement φ _{w of the particles 802 in FIG. 8 (a).} Can be done. η _t is a viscosity constant in the shear direction and is preset. Δt is a predetermined time interval and is set in advance.
The Young's modulus and Poisson's ratio set for the wall particles as attribute information of the structure (swivel chute in this embodiment) are K _n , η _n shown in Eq. (2) and Eq. (3). It is used in determining K _t and η _{t to be indicated.}

In equation (4), v is the velocity of the particles for analysis (see FIG. 8). Attached above v represents the time derivative. F is the sum of the repulsive forces in the shear direction and the normal direction. m is the mass of the particles for analysis and is preset. g is the gravitational acceleration. When the analysis particles are not in contact with other analysis particles or wall particles, the first term on the right side of Eq. (4) becomes 0 (zero).
In equation (5), ω is the angular velocity of the particles for analysis (see FIG. 8). Attached above ω represents the time derivative. M is the sum of the moments of force and is obtained from _{the repulsive forces F n, sw} , F _{t, sw in the shear direction and the normal direction.} I is the moment of inertia, which is preset.

The method for analyzing the behavior of the particles for analysis (the analysis method executed by the particle behavior analysis unit 108 for analysis) is the method described in Patent Document 1, but there are various aspects of the behavior of the particles in DEM. The known analysis method described above can be applied, and the method is not limited to the methods using the above-mentioned equations (2) to (5).
Further, in the above embodiments, an example of analyzing the behavior of spherical analysis particles by DEM has been described, but the shape of the analysis particles is not limited to this. That is, the analysis particles may be non-spherical particles, and for example, as described in Patent Document 2 (Japanese Unexamined Patent Publication No. 2011-69656), a polyhedron having a spherical contact region at each apex is used for analysis. As the particles, non-spherical particles may be reproduced.
FIG. 9 is a diagram showing an example of the behavior of particles analyzed by DEM. FIG. 9 shows the state of particles at a certain time.

(Particle behavior display unit 109 for analysis)
The analysis particle behavior display unit 109 displays the behavior of each analysis particle analyzed by the analysis particle behavior analysis unit 108 on the computer display.

Next, an example of the processing of the particle behavior analysis device 100 when setting the wall particles will be described with reference to the flowchart of FIG.
First, in step S1001, the point cloud data acquisition unit 101 acquires the point cloud data.
Next, in step S1002, the point data removing unit 102 determines whether or not there is a point to be removed in the point cloud data acquired in step S1001. As a result of this determination, if there is a point to be removed in the point cloud data acquired in step S1001, the process proceeds to step S1003, and if not, the process proceeds to step S1004, which will be described later, omitting step S1003. ..

When the process proceeds to step S1003, the point data removing unit 102 identifies the point to be removed and removes the data of the specified point from the point cloud data. The point cloud data not removed in step S1003 is the analysis target candidate point cloud data.
Then, when the process proceeds to step S1004, the point data complementing unit 103 determines whether or not there is a point complementing region in the region where the analysis target candidate point cloud data exists. As a result of this determination, if there is an area that complements the points in the area where the analysis target candidate point cloud data exists, the process proceeds to step S1005, and if not, the process skips step S1005 and proceeds to step S1006 described later. move on.

When the process proceeds to step S1005, when the point data complementing unit 103 determines that there is an area for complementing the points, the point data is added to the area, and the point data is added to the analysis target point.
Then, when the process proceeds to step S1006, the point cloud data extraction unit 104 arranges the wall particles from the analysis target candidate point cloud data so that the distribution of the spatial frequency of the analysis target points has a blue noise characteristic. (Analysis target point) is extracted.

Next, in step S1007, the maximum value (maximum surface roughness value ζ _{i, j} ) of the amount of dent from the surface of the structure (swivel chute in this embodiment) is acquired as surface roughness information.
Next, in step S1008, the wall particle setting unit 106 sets wall particles one by one for each of the analysis target points extracted in step S1006 based on the surface roughness information acquired in step S1007. Set.
Next, in step S1009, the attribute information of the structure (swivel chute in this embodiment) is acquired for each of the wall particles set in step S1008.
After the process of step S1009, the process according to the flowchart of FIG. 10 ends.

Next, an example of processing of the particle behavior analysis device 100 when analyzing the behavior of the particles for analysis will be described with reference to the flowchart of FIG. Here, it is assumed that the particles for analysis are already arranged.
First, in step S1101, the analysis particle behavior analysis unit 108 sets the analysis time t to Δt. This Δt corresponds to Δt in Eqs. (2) and (3).
Next, in step S1102, the analysis particle behavior analysis unit 108 selects one analysis particle.
Next, in step S1103, the analysis particle behavior analysis unit 108 determines whether or not the analysis particles selected in step S1102 are in contact with the wall particles. As a result of this determination, if the analysis particles are in contact with the wall particles, the process proceeds to step S1104. On the other hand, when the analysis particles are not in contact with the wall particles, the process skips step S1104 and proceeds to step S1105 described later.

When the process proceeds to step S1104, the analysis particle behavior analysis unit 108 applies the force (repulsive force in the shear direction and the normal direction) acting on the analysis particle selected in step S1102 from the wall particle to the equation (2) and The velocity v of the particles for analysis is calculated by the equation (4), and the angular velocity ω of the particles for analysis is determined by the equation (5) based on the result. At this time, it is assumed that no force acts from the analysis particles to the wall particles.
Then, when the process proceeds to step S1105, the analysis particle behavior analysis unit 108 determines whether or not the analysis particles selected in step S1102 are in contact with other analysis particles. As a result of this determination, if the analysis particles are in contact with other analysis particles, the process proceeds to step S1106. On the other hand, when the analysis particles are not in contact with other analysis particles, the process skips step S1106 and proceeds to step S1107 described later.

When the process proceeds to step S1106, the analysis particle behavior analysis unit 108 applies a force (repulsive force in the shear direction and the normal direction) acting on the analysis particle selected in step S1102 from the other analysis particles (2). The velocity v of the analysis particle is determined by the equation (4), and the angular velocity ω of the analysis particle is determined by the equation (5) based on the results.
Then, when the process proceeds to step S1107, the analysis particle behavior analysis unit 108 determines whether or not all the analysis particles have been selected. As a result of this determination, if all the analysis particles have not been selected, the process returns to step S1102, and the analysis particle behavior analysis unit 108 selects one unselected analysis particle.

On the other hand, when all the analysis particles are selected, the process proceeds to step S1108. When the process proceeds to step S1108, the analysis particle behavior analysis unit 108 moves the analysis particles based on the calculation results of steps S1104 and S1106.
Next, in step S1109, the analysis particle behavior analysis unit 108 determines whether or not the analysis time t has reached the analysis end time T. The analysis end time T is set in advance based on, for example, an input operation by the operator to the particle behavior analysis device 100.
As a result of this determination, if the analysis time t is not the analysis end time T, the process proceeds to step S1110, and the analysis particle behavior analysis unit 108 advances the analysis time t by Δt to update the analysis time t. Then, the process returns to step S1102, and the behavior of each analysis particle at the updated analysis time t is obtained.
On the other hand, when the analysis time t becomes the analysis end time T, the process proceeds to step S1111, and the analysis particle behavior display unit 109 performs each analysis from the analysis start time (analysis time t = 0) to the analysis end time T. The behavior of the particles is displayed on the display. After the process of step S1111, the process according to the flowchart of FIG. 11 ends.

Here, the reason why it is preferable that the distribution of the spatial frequency of the analysis target point has the blue noise characteristic will be described.
FIG. 12 is a diagram showing another example of arrangement of analysis target points. As shown in FIG. 12, the analysis target points are extracted from the analysis target candidate point cloud data at equal intervals along the x-axis, y-axis, and z-axis in the three-dimensional Cartesian coordinate system, and the analysis target points are centered. When the spherical particles to be arranged are arranged as wall particles, the wall particles are regularly arranged. In this case, the direction along the orientation of the wall particles (the direction along the x-axis, y-axis, or z-axis in the three-dimensional Cartesian coordinate system (for example, the direction of the arrow line 1201)) and the orientation of the wall particles are not followed. There is a possibility that the flow of the analysis particles (raw material) may change depending on the direction (for example, the direction of the arrow line 1202).

When the particle size of the analysis particles is smaller than the particle size (diameter) of the wall particles, the flow of the analysis particles is greatly affected by the orientation of the wall particles. This is because the analysis particles can easily enter the region generated between the wall particles adjacent to each other. Therefore, if the particle size of the wall particles is made smaller than the particle size of the analysis particles, the flow of the analysis particles is less likely to be affected by the orientation of the wall particles. However, if the particle size of the wall particles is reduced, the number of wall particles increases. Further, when the difference between the particle size of the wall particles and the particle size of the analysis particles becomes large, it is necessary to shorten the time interval Δt when analyzing the behavior of the analysis particles by DEM, so that the calculation time increases. .. Therefore, the particle size of the wall particles is preferably within the range of the particle size distribution of the raw material (the diameter distribution when the shape of the raw material is approximated by a sphere). The particle size of about 80 [%] of the minimum particle size when the shape of the raw material is approximated by a sphere is the minimum particle size of the wall particles that can be calculated in a realistic calculation time. For example, when analyzing the behavior of a raw material having a particle size (diameter) of 15 [mm] to 55 [mm], the particle size (diameter) of the wall particles needs to be, for example, 12 [mm] or more. When wall particles having such a particle size are arranged at equal intervals along the x-axis, y-axis, or z-axis in a three-dimensional Cartesian coordinate system, the analysis particles having a particle size of 15 [mm] are for walls. It is greatly affected by the orientation of the particles. Therefore, there is a risk that the behavior of the analysis particles cannot be analyzed accurately.

As described above, if the wall particles are regularly aligned, the flow of the analysis particles may change between the direction along the orientation of the wall particles and the direction not along the orientation of the wall particles. be. Further, if the spacing between the wall particles is not uniform, the unevenness of the surface of the structure (in this embodiment, the swirling chute) will differ depending on the location, and the friction between the wall particles and the analysis particles may change. There is.
In order to suppress these things, in the present embodiment, the variation in the distances between the analysis target points adjacent to each other is small, and the analysis target points are arranged isotropically (that is, not biased in a specific direction). As described above, the spatial frequency distribution of the analysis target points is made to have a blue noise characteristic in which the points can be arranged periodically and isotropically.

Next, the method of this embodiment is compared with the method described in Patent Document 1.
Point cloud data on the surface of the swirl chute was measured using a three-dimensional laser scanner. As a result, data of 278,554 points was obtained. Using the same computer (computer with the same hardware such as CPU and memory), wall particles were generated by each of the method described in this embodiment and the method described in Patent Document 1. Here, for comparison, 100,000 wall particles were generated by either method. Also, in each method, wall particles were placed over the entire area of the swirl chute. In addition, the method of this embodiment does not complement the data.

As a result, in the method described in Patent Document 1, it takes 10.62 [seconds] to create the surface information (mesh) of the swirling chute from the point cloud data, and 2.71 to arrange the wall particles. [Second] It took. That is, in the method described in Patent Document 1, the time required from the acquisition of the point cloud data to the setting of the wall particles was 13.31 [seconds].
On the other hand, in the method of the present embodiment, the time required to generate the analysis target point cloud data by thinning out the point cloud data was 0.47 [seconds]. As described above, in this example, by using the method of this embodiment, the time required to set the wall particles can be reduced to about 1/28 as compared with the method described in Patent Document 1. This difference is expected to increase as the number of point cloud data increases.

As described above, in the present embodiment, the particle behavior analysis device 100 extracts the analysis target points from the point cloud data obtained by the three-dimensional measurement so that the spatial frequency distribution has the blue noise characteristic, and the extracted analysis target points. Set the wall particles for. Therefore, it is not necessary to use the surface shape of the structure (swivel chute in this embodiment) in order to arrange the wall particles. Therefore, the wall particles can be arranged without increasing the calculation load. Further, in the present embodiment, since the distribution of the spatial frequency of the analysis target point has a blue noise characteristic, for example, even when analyzing the behavior of particles whose flow direction is not fixed, the structure (in the present embodiment). The surface of the swirl chute) can be accurately represented by wall particles, and calculations for analyzing the behavior of the particles can be performed with high accuracy without increasing the calculation load.

Further, in the present embodiment, the particle behavior analysis device 100 removes points in the region that do not contribute to the flow of particles from the point cloud data obtained by the three-dimensional measurement. Therefore, it is possible to suppress the generation of unnecessary wall particles. Therefore, the calculation load when arranging the wall particles can be further reduced.
Further, in the present embodiment, the particle behavior analysis device 100 adds points to a region having a low density of points among the regions contributing to the flow of particles. Therefore, for example, even if there is a region where it is difficult to obtain points by three-dimensional measurement, the surface of the structure (swivel chute in this embodiment) is accurately represented by wall particles over the entire region that contributes to the flow of particles. can do.

(Modification example)
In the present embodiment, when the point data complementing unit 103 determines that there is a region to complement the points in the region where the analysis target candidate point cloud data exists, the point data is added to the region as an example. I mentioned and explained. However, it is not always necessary to do this. For example, the particle behavior analysis device 100 has a wall surface information derivation unit instead of the point data complement unit 103, and the wall surface information derivation unit creates a mesh based on the point data included in the region that complements the points. The information of the created mesh may be set as the wall surface information (information indicating the outline of the shape of the surface of the structure) described in Patent Document 1. In this case, the wall particle setting unit 106 describes the region of the surface of the structure (swivel chute in the present embodiment) in which the wall surface information (mesh) is generated, as described in Patent Document 1. The wall particles are set in this way.

Explaining a specific example, for the wall particles set in the region where the wall surface information (mesh) is generated, information on _{the radius r w of the wall particles is obtained as described in Patent Document 1.} Give. _{Then, the wall particle setting unit 106 is arranged inside the structure and has a radius r w} at both ends of a region that can be approximated to a straight line on the surface of the structure so as to be in contact with the surface (mesh) of the structure. The wall particles to be held are set as reference wall particles (reference wall particles). Further, the wall particle setting unit 106 mutually uses _{the maximum value ζ i, j of} the surface roughness corresponding to the region where the wall surface information (mesh) is generated _{and the radius r w} of the wall particles. The distance P between the centers of the wall particles adjacent to is calculated. Then, the wall particle setting unit 106 is arranged inside the structure with a distance P between the centers between the adjacent wall particles, and is in contact with the surface (mesh) of the structure. As described above, the wall particles _{having a radius r w} are set between the reference wall particles.

Further, in the present embodiment, a case where spherical particles centered on the analysis target point are set as wall particles has been described as an example. However, this is not always necessary if the wall particles are set for the analysis target point. For example, the wall particle setting unit 106 is one of the surfaces of the wall particles so that the position on the analysis target point among the positions of the surface of the wall particles is the position closest to the surface side of the structure. The wall particles may be set by adjusting the position to the position of the analysis target point.

Further, in the present embodiment, the spatial frequency distribution of the analysis target points has a blue noise characteristic. By doing so as described above, for example, even when analyzing the behavior of particles whose flow direction is not fixed, the surface of the structure (swivel chute in the present embodiment) can be accurately represented by wall particles. It is preferable because it can be done. However, this is not always necessary as long as the variation (variance) in the distances of the analysis target points adjacent to each other is smaller than the variation (dispersion) of the points before the analysis target points are extracted. For example, when analyzing the behavior of a particle whose flow direction is limited to a specific one direction, the point cloud data extraction unit 104 uses three-dimensional Cartesian coordinates from the analysis target candidate point cloud data as shown in FIG. The analysis target points may be extracted at equal intervals along the x-axis, y-axis, and z-axis in the system. In this case, the arrangement of the plurality of points added by the point data complementing unit 103 is preferably determined by the same guideline as the guideline for determining the arrangement of the plurality of analysis target points. In the example described here, the point cloud data complement unit 103 adds points in the mesh at equal intervals along the x-axis, y-axis, and z-axis in the three-dimensional Cartesian coordinate system. As described above, the point cloud data extraction unit 104 may add points at random without doing this, for example.

Further, in the present embodiment, a part for setting wall particles (point cloud data acquisition unit 101, point data removal unit 102, point data complement unit 103, point cloud data extraction unit 104, wall particle setting unit 106). Was included in the particle behavior analysis device 100 as an example. However, the part for setting the wall particles (point cloud data acquisition unit 101, point data removal unit 102, point data complement unit 103, point cloud data extraction unit 104, wall particle setting unit 106) is a particle behavior analysis device. It may be arranged in a device different from 100. In this case, the device functions as a structure model generation device that generates a model of the structure (models the structure) using the wall particles.

The embodiment of the present invention described above can be realized by executing a program by a computer. Further, a computer-readable recording medium on which the program is recorded and a computer program product such as the program can also be applied as an embodiment of the present invention. As the recording medium, for example, a flexible disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a magnetic tape, a non-volatile memory card, a ROM, or the like can be used.
In addition, the embodiments of the present invention described above are merely examples of embodiment of the present invention, and the technical scope of the present invention should not be construed in a limited manner by these. It is a thing. That is, the present invention can be implemented in various forms without departing from the technical idea or its main features.

100: Particle behavior analysis device, 101: Point group data acquisition unit, 102: Point data removal unit, 103: Point data complement unit, 104: Point group data extraction unit, 105: Surface roughness information acquisition unit, 106: For walls Particle setting unit, 107: Attribute information setting unit, 108: Analysis particle behavior analysis unit, 109: Analysis particle behavior display unit, 601-602: Analysis target point, 603 to 604: Wall particles

Claims (14)

It is a structure model generator that uses a plurality of wall particles to generate a model of the structure when analyzing the behavior of a plurality of analysis particles in the region where the structure exists by the discrete element method. hand,
A point cloud data acquisition means for acquiring data including the positions of a plurality of points indicating positions on the surface of the structure as point cloud data, which is data measured by a three-dimensional measuring device.
A point cloud data extraction means for extracting a part of a plurality of points included in the point cloud data as analysis target points and extracting data including the positions of the plurality of analysis target points as analysis target point cloud data.
Using the analysis target point cloud data, the wall particle setting means for setting the wall particles for each of the plurality of analysis target points included in the analysis target point cloud data is provided.
The point cloud data extraction means has a smaller variation in the distances between adjacent points at the plurality of analysis target points than a variation in the distances between the plurality of points included in the point cloud data. A structure model generator, characterized in that data including the positions of the plurality of analysis target points is extracted as analysis target point cloud data. The point cloud data extraction means extracts data including the positions of the plurality of analysis target points as analysis target point cloud data so that the spatial frequency distribution of the plurality of analysis target points has blue noise characteristics. The structure model generation device according to claim 1. Point data removal that removes data indicating the position of a point included in a region specified according to the shape of the structure and the path of the analysis particle from the point cloud data acquired by the point cloud data acquisition means. Have more means,
The point cloud data extracting means is a part of a plurality of points included in the point cloud data not removed by the point data removing means among the point cloud data acquired by the point cloud data acquiring means. The structure model generation apparatus according to claim 1 or 2, wherein the data including the positions of a plurality of the analysis target points is extracted as the analysis target point cloud data. A plurality of points are added in a region specified according to the density of points in the point cloud data acquired by the point cloud data acquisition means, and data including the positions of the added plurality of points is included in the point cloud data. The structure model generation device according to any one of claims 1 to 3, further comprising a point data complementing means. The structure model generation according to claim 4, wherein the point data complementing means determines the arrangement of the plurality of points to be added by the same guideline as the guideline for determining the arrangement of the plurality of analysis target points. Device. The position of a point in the point cloud data acquired by the point cloud data acquisition means and included in a region specified according to the density of the points in the point cloud data acquired by the point cloud data acquisition means. Further having a wall surface information deriving means for deriving wall surface information indicating an outline of the shape of the surface of the region based on the information.
The wall particle setting means does not use the point cloud data to be analyzed for a region specified according to the density of points in the point cloud data acquired by the point cloud data acquisition means, and the wall is not used. Any one of claims 1 to 3, wherein the wall particles are arranged inside the surface of the structure indicated by the surface information and are in contact with the surface of the structure. The structure model generator described in 1. Further, it has a surface roughness information acquisition means for acquiring surface roughness information indicating the maximum value of the amount of dent from the surface of the structure.
In the wall particle setting means, the maximum value of the dent amount of the gap formed between the wall particles arranged adjacent to each other is the dent amount from the surface of the structure shown in the surface roughness information. The structure model generation device according to any one of claims 1 to 6, wherein the wall particles are set so as to have a value corresponding to the maximum value. The wall particle setting means sets the size of the wall particles with respect to the analysis target point based on the distance between the analysis target points arranged adjacent to each other and the surface roughness information. The structure model generation device according to claim 7, wherein the structure model is calculated. The wall particle setting means derives a representative value of the distance between two analysis target points arranged adjacent to each other, and indicates the representative value and the representative value of the amount of dent from the surface of the structure. Any one of claims 1 to 6, wherein the size of the same wall particles set for each of the plurality of analysis target points is calculated based on the surface roughness information. The structure model generator described in 1. A particle behavior analysis means for analysis that analyzes the behavior of a plurality of particles for analysis in a region where a structure exists by a discrete element method, and a structure model generation device according to any one of claims 1 to 9. It is a particle behavior analysis device having
When the analysis particle behavior analysis means determines that the analysis particles have come into contact with the wall particles set by the wall particle setting means included in the structure model generation device, the wall particles are described as described above. A particle behavior analysis device characterized in that a force received by an analysis particle is applied to the analysis particle to move the analysis particle. This is a structure model generation method that uses a plurality of wall particles to generate a model of the structure when analyzing the behavior of a plurality of analysis particles in the region where the structure exists by the discrete element method. hand,
Point cloud data acquisition means that the point cloud data acquisition means is data measured by a three-dimensional measuring device and acquires data including the positions of a plurality of points indicating the positions on the surface of the structure as point cloud data. Process and
The point cloud data extraction means sets some points among the plurality of points included in the point cloud data as analysis target points, and extracts data including the positions of the plurality of analysis target points as analysis target point cloud data. Point cloud data extraction process and
A wall particle setting step in which the wall particle setting means sets the wall particles for each of the plurality of analysis target points included in the analysis target point cloud data using the analysis target point cloud data. And have
In the point cloud data extraction step, the variation in the distances between the points to be analyzed at the plurality of analysis target points is smaller than the variation in the distances between the points adjacent to each other at the plurality of points included in the point cloud data. A method for generating a structure model, which comprises extracting information including the positions of a plurality of analysis target points as analysis target point cloud data. The particle behavior analysis step for analysis in which the particle behavior analysis means for analysis analyzes the behavior of a plurality of particles for analysis in the region where the structure exists by the discrete element method, and the structure model generation according to claim 11. It is a particle behavior analysis method having each step of the method.
When the analysis particle behavior analysis step determines that the analysis particles are in contact with the wall particles set in the wall particle setting step in the structure model generation method, the analysis is performed from the wall particles. A particle behavior analysis method characterized in that a force received by a particle is applied to the analysis particle to move the analysis particle. A program characterized in that a computer functions as each means of the structure model generator according to any one of claims 1 to 9. A program characterized in that a computer functions as each means of the particle behavior analysis apparatus according to claim 10.

Images