JP2010152863A - System and method for extracting boundary node, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce processing time of node extraction, and perform, for assembly components, contact determination between components in a short time. <P>SOLUTION: A function block of a central processing unit (shown in a different drawing) includes: an input means 11 which inputs analysis model data; an area radius setting means 12 which sets an existing area of nodes in extraction of boundary nodes from the data; a threshold setting means 13 which sets two thresholds for discriminating the boundary nodes; an area internal node extraction means 14 which extracts nodes existing in a circle having one of the set thresholds as a radius; a vector calculation means 15 which performs vector calculation for a group of extracted nodes; a boundary node determination means 16 which determines boundary nodes based on the vector calculation result and the other of the set thresholds; a same node-sharing element search means 17 which searches elements sharing one node; and an output means 18 which outputs information for the extracted nodes. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a boundary node extraction system and a boundary node extraction method, and in particular, a boundary node that efficiently extracts nodes existing at the boundary of an analysis model used for numerical analysis using a finite element method or the like. The present invention relates to an extraction system, a boundary node extraction method, and a program.

Traditionally, there are many cases where quality problems due to design defects have evolved into serious incidents related to the existence of companies, so design quality verification methods and quality improvement have become important issues. Yes. Under such circumstances, as a design quality verification method, an analysis method using a finite element method or the like is used in many fields.
Recently, with the spread of three-dimensional CAD, design drawings are three-dimensionalized and developed as three-dimensional data. In addition, an analysis model of the finite element method is created based on such three-dimensional data, and analysis calculation is executed. In recent years, with the remarkable improvement in PC (personal computer) processing capability, it is possible to analyze an analysis model that faithfully reproduces the actual shape of the object to be analyzed in a three-dimensional space, or a structure model in which multiple parts are assembled. As a result, the analysis model of the finite element method tends to increase in scale and the number of elements tends to increase.

  A structure composed of a plurality of parts realizes the transmission of force between parts and the restraint of displacement by considering the presence or absence of contact possibility. When determining whether or not to consider the possibility of contact, a pair of two parts is extracted from a plurality of parts, and nodes are extracted from each part of the pair. Calculate the distance between any two nodes. Note that these operations are performed by combining all the nodes of both parts forming the pair, thereby obtaining the minimum distance between the two nodes. Next, this minimum distance is compared with a predetermined threshold for determining whether or not to consider the possibility of contact, and both parts can touch unless this minimum distance exceeds this threshold. The presence or absence of contact between both parts is determined as having the property.

The above determination process becomes enormous as the number of parts increases because the number of combinations of the number of pairs to be considered increases at a rate of about the square of the number of parts. Similarly, the number of node combinations for calculating the distance also increases at a ratio of the square of the number of nodes, and becomes enormous as the number of nodes increases.
However, the combination of nodes in the calculation for obtaining the minimum distance between parts may actually be a combination of only the nodes existing at the boundary. Therefore, in the conventional processing for all the internal nodes, there are many combinations. Therefore, it was a problem to solve this point. Therefore, as a technique for solving such problems, there has been a strong demand for the development of a technique for efficiently extracting boundary nodes from an analysis model.

  Conventionally, as a boundary node extraction method for solving the above-mentioned problem, for example, in the boundary surface extraction method described in Patent Document 1, a table representing the coordinates of the nodes and the nodes constituting the elements is created, Take out one element, specify the faces that make up this element, specify the nodes that make up this face, and then take out the elements one by one and check the constituent nodes of the faces that make up the element In the case where the same object does not exist, the boundary surface is used.

Further, for example, in the node extraction method described in Patent Document 2, first, all the surfaces constituting the element are used as boundary surfaces, and the elements are sequentially extracted one by one as in the case of Patent Document 1 described above, The boundary surface is extracted by checking whether or not there are identical nodes on the surfaces constituting the element.
However, this method has a problem that the processing time becomes enormous because it is necessary to check whether or not the nodes are configured with the same node in all combinations of the constituent surfaces of all elements.

FIG. 16 is a flowchart showing the processing procedure of the conventional boundary node extraction system.
Hereinafter, the node extraction method will be described in the case of the cubic analysis model shown in FIG. 6 with reference to the processing procedure of the conventional boundary node extraction method shown in FIG.
In the prior art, a boundary node is extracted by paying attention to the fact that the surface existing at the boundary is not shared with the surfaces of other elements. The number of elements of the rectangular parallelepiped model shown in FIG. 6 is 1000, and the number of nodes is 1331.
As can be seen from FIG. 16, in the prior art, the contact table is created in step A1, the element table is created in step A2, and then the following three loops (repetition processing, Loop1,2,3) are executed.

(1) Loop1 is a loop for selecting an element from the step (step A3) for selecting the element a to the final determination condition (step A10).
(2) Loop2 is a loop for selecting an element from the selection of element b (step A4) to the final determination condition (step A9).
(3) Loop3 extracts the constituent surface of each element (step A5), checks whether or not they are the same surface (step A6), and identifies them if they are the same surface (step A7) A loop that repeats until the determination condition (step A8) for determining whether or not the survey has been completed.

  Each Loop has a nested structure of Loop1⊃Loop2⊃Loop3. In order to process elements b1 to bj for one element ai, a maximum of 1000 × 1000 = 1000000 times of repetition processing is executed with a nested structure of Loop1 and Loop2. If duplication is excluded, 1000 × 999/2 = 499500 iterations will be performed. Furthermore, in the case of a hexahedral element in the process of Loop 3, 6 × 6 = 36 checks of the same surface are required, and the same surface identification process (step A7) is performed 499500 × 36 = 17982000 times. .

JP-A-8-263697 Japanese Patent Laid-Open No. 10-289257

However, in the conventional boundary node extraction system and boundary node extraction method described in the background art above, whether or not all the constituent surfaces and sides of all elements of the analysis model share the same surface and side. However, there is a problem that the processing time becomes enormous.
In addition, in an analysis model of a structure composed of a plurality of parts, when the contact determination between parts is determined by the minimum distance between two parts, contact is made between all nodes in each combination of all parts. Since the determination distance calculation is performed, there is a problem that the processing time is enormous in this processing step.

The present invention has been made in view of the above-described conventional problems, and is a boundary node extraction capable of reducing the processing time required to extract a node existing at the boundary of a given analysis model as a processing target. It is an object to provide a system, a boundary node extraction method, and a program.
Another object of the present invention is to extract a boundary node for a single component, and when assembling the component, it is possible to search for the minimum distance between components only with the boundary node, thereby shortening the contact determination between the components. It is an object to provide a boundary node extraction system, a boundary node extraction method, and a program that can be performed in time.

  In order to solve the above-described problem, a boundary node extraction system according to the present invention is a boundary node extraction system that extracts nodes existing at the boundary of an analysis model used in numerical analysis, and the nodes of the nodes constituting the analysis model are extracted. Based on position information, for each of the nodes, an in-region nodal point extracting means for extracting other nodal points existing in a circle centered on the nodal point and having a predetermined area radius as a radius, and the center of the circle A position vector constructing means for creating a position vector having the above-mentioned node as the tail and the extracted other node as a tip, and a combined vector calculating means for obtaining a combined vector of the position vectors and calculating the magnitude of the vector Boundary node extraction means for extracting the node as the center of the circle as a candidate for a boundary node when the size of the combined vector exceeds a predetermined first threshold; There is provided a boundary node extraction system, characterized in that there was e.

  The boundary node extraction method is a boundary node extraction method for extracting nodes existing at the boundary of an analysis model used in numerical analysis, and is based on position information of nodes constituting the analysis model, for each of the nodes. , A node extraction step in a region that extracts other nodes existing in a region having a radius of a preset region radius, and a node at the center of the circle as a tail, A position vector constructing step for creating a position vector having the extracted other node as a tip; a combined vector calculating step for obtaining a combined vector of the position vector and calculating a magnitude of the vector; and a size of the combined vector A boundary node extracting step of extracting the node at the center of the circle as a candidate for a boundary node when a predetermined first threshold is exceeded. .

  Further, the computer program according to the present invention provides a computer that extracts the nodes existing at the boundary of the analysis model used in the numerical analysis, and sets the nodes for each of the nodes based on position information of the nodes constituting the analysis model. A node extraction step in a region that extracts other nodes existing in a region having a radius of a preset region radius as a center, and a node that is the center of the circle as a tail, and the extracted other A position vector constructing step for creating a position vector having a node as a tip; a composite vector calculating step for obtaining a composite vector of the position vector and calculating a magnitude of the vector; and a magnitude of the composite vector being a predetermined first A boundary node extraction step of extracting the node at the center of the circle as a boundary node candidate when the threshold of And butterflies.

  As described above, according to the boundary node extraction system of the present invention, since the combination of the constituent surfaces of all elements can be eliminated and the nodes existing at the boundary can be extracted by the search of only the nodes, It is possible to reduce the time required to extract the nodes to be performed, and since the boundary node is extracted by the part alone and the minimum distance search is performed by combining only the boundary nodes extracted by the part unit after the parts assembly, There is an effect that it is possible to determine an efficient contact condition (that is, whether there is a possibility of contact).

It is a block diagram which shows the whole structure of the boundary node extraction system which concerns on embodiment of this invention. It is a functional block diagram which shows the functional block structure in the central processing unit 3 of the boundary node extraction system which concerns on embodiment of this invention. It is a flowchart figure which shows the best process sequence of the boundary node extraction system which concerns on embodiment of this invention. It is another flowchart figure which shows the best process sequence of the boundary node extraction system which concerns on embodiment of this invention. It is another flowchart figure which shows the best process sequence of the boundary node extraction system which concerns on embodiment of this invention. An example of a three-dimensional analysis model is shown. An example of an ideal analysis model for explaining a vector calculation method for extracting boundary node candidates is shown. It is explanatory drawing which shows the display screen as an example of the analysis model in the boundary node extraction system which concerns on embodiment of this invention. It is explanatory drawing which listed one example of the node in the area | region extracted corresponding to each node. It is explanatory drawing explaining a node sharing element. It is explanatory drawing which shows the relationship between the composite vector P of the position vector p which makes the end point the node in the area | region at the time of extracting the candidate of a boundary node, and a threshold value. It is explanatory drawing which shows the state which completed the boundary node extraction with respect to the analysis model shown in FIG. An example of an analysis model is shown. It is explanatory drawing which showed the result of having analyzed the distance between nodes of a some analysis model. It is a table | surface which shows the result implemented by making a contact determination threshold value into 0.3 mm. It is a flowchart figure which shows the process sequence of the conventional boundary node extraction system. It is a figure which shows the system display state of the analysis model of the Example of this invention. It is a figure which shows the process outline | summary of the node extraction in an area | region of the Example of this invention. It is a figure which shows the area | region node table of the Example of this invention. It is a figure which shows the boundary node extraction completion state of the Example of this invention.

  According to the analysis principle of the boundary node extraction system of the present invention, first, attention is paid to a certain node of the analysis model, and a node existing in a region centered on the node is extracted. Next, a position vector of the extracted node with respect to the focused node is obtained, and a combined vector of all position vectors is obtained. Next, when the size of the combined vector is larger than the threshold value, it is possible to efficiently extract the boundary node by determining that the possibility of being the boundary node is high. In addition, the above processing is performed on a single part body, boundary nodes are extracted, and when the parts are assembled and contact judgment is performed, only the extracted boundary nodes are used to search for the minimum distance between parts. Thus, it is possible to efficiently determine the contact condition (that is, whether there is a possibility of contact).

The best mode of the boundary node extraction system of the present invention will be described below in detail with reference to the drawings.
FIG. 1 is a configuration diagram showing the overall configuration of a boundary node extraction system according to an embodiment of the present invention.
The boundary node extraction system shown in the figure includes an input device 1 that can be configured with a keyboard, a mouse, etc., a data processing device 2 that stores and executes information necessary for boundary node extraction processing, and boundary node extraction processing. An external storage device 6 that stores a large amount of necessary information, a display device 7 such as a display device, and an output device 8 such as a printing device are provided. The data treatment device 2 includes a central treatment device 3 that executes information processing, a main storage device 4 that temporarily stores data being processed in the central treatment device 3 and a processing program necessary for boundary node extraction processing; An auxiliary storage device 5 that permanently stores information stored in the built-in HDD or the like.

FIG. 2 is a functional block diagram showing a functional block configuration in the central processing unit 3 of the boundary node extraction system according to the embodiment of the present invention.
The boundary node extraction system according to the embodiment of the present invention is an analysis model as a functional block provided in the central processing unit 3 via the input device 1 or from the auxiliary storage device 5 or the external storage device 8 as shown in FIG. An input unit 11 for capturing data, an area radius setting unit 12 for setting an existing region of nodes to be considered when extracting boundary nodes from the captured analysis model data, and two threshold values for determining boundary nodes are set. And a threshold setting means 13 for extracting the nodes existing in a circle centered on the node and having a radius of the one of the set thresholds, and an intra-area node extracting means 14 (intra-area node extracting means). Vector calculation means 15 (position vector construction means, combined vector calculation means) for performing vector calculation with the node group, the vector calculation result, and the other set A boundary node determination unit 16 (boundary node extraction unit) that determines a boundary node based on a value, an identical node shared element search unit 17 (shared element count unit, boundary node determination unit) that searches for an element that shares one node; Output means 18 for outputting the extracted node information to the display device 7 and the output device 8.

  The functional blocks shown in FIG. 2 are configured as an executable computer program, and the executable computer program is stored in the external storage device 6 or the auxiliary storage device 5 shown in FIG. When executing the extraction of boundary nodes, the computer program is started from the input device 1, and this computer program is called into the main storage device 4. Although the computer program is operated via the input device 1, the boundary node extraction processing is executed by the central processing device 3, and the execution result is output to the display device 7. The execution result can also be stored in the external storage device 6 or the auxiliary storage device 5.

Hereinafter, the function of the boundary node extraction system according to the embodiment of the present invention will be described.
FIG. 3 is a flowchart showing the best processing procedure of the boundary node extraction system according to the embodiment of the present invention.
Hereinafter, the best processing procedure of the boundary node extraction method according to the embodiment of the present invention will be described using the flowchart shown in FIG. 3 with reference to FIGS.
(Step S1)
First, in step S1, the input means 11 imports analysis model data into the boundary node extraction system, extracts node numbers and node coordinate value data from the imported analysis model data, and creates a node table.

(Step S2)
Next, in step S2, the input means 11 extracts element numbers and node numbers constituting the elements of the element numbers from the imported analysis model data, and creates an element table. The imported analysis model data is created in accordance with an original format that can be interpreted by an analysis solver (each means excluding the input means 11 from the functional block shown in FIG. 2) that executes the analysis calculation. Thereafter, the area radius setting means 12 (FIG. 2) sets an area radius considering the position vector for each node of the imported analysis model, and the threshold setting means 13 (FIG. 2) sets the position vector of each position vector. It is assumed that a threshold value (first threshold value) for the size of the combined vector and a threshold value (second threshold value) for the number of shared elements are set. The region radius and the threshold value can be set as an integer multiple of the basic element size by first obtaining the average element size as the basic element size based on the data of the analysis model.

(Step S3)
Next, in step S3, the in-region node extracting means 14 selects one node from the imported node data. The nodes are selected from the node having the smallest node number (= 1) to the number of contacts (= M) sequentially. This node number is normally given sequentially by a natural number. Therefore, in the i-th execution of step S3, the node i whose node number is i is selected.
(Step S4)
Next, in step S4, the in-region node extracting means 14 extracts the nodes existing within the area radius preset by the area radius setting means 12 with the node i selected as described above as the center. Create a table.
(Step S5)
Next, in step S5, the vector calculation means 15 calculates the position vector p of each node in the in-region node table for the node i. The position vector p centered on the node i can be drawn by the number of nodes in the in-region node table.

(Step S6)
Next, in step S6, the vector calculation means 15 calculates a combined vector P of the vector p, and calculates the magnitude | P | of the combined vector.
(Step S7)
Next, in step S7, the boundary node determination unit 16 compares the threshold set by the threshold setting unit 13 with the size of the combined vector | P |, and the size of the combined vector | P | Is less than the threshold (in the case of NO), the process proceeds to step S11. Otherwise, if the magnitude | P | of the combined vector is equal to or larger than the threshold (in the case of YES), the process proceeds to step S8.
In this case, it is considered that there is a high possibility that the node i exists at the boundary (a position that is a boundary surface between the inside and the outside of the shape indicated by the analysis model data). Hereinafter, this case will be described in more detail.

FIG. 7 shows an example of an ideal analysis model for explaining a vector calculation method for extracting boundary node candidates.
Describing with an ideal cuboid analysis model formed of cuboid elements shown in FIG. 6, FIG. 7 shows a plan view seen from a direction perpendicular to a surface having the cuboid. Then, in the case of an ideal analysis model formed by square elements as shown in FIG. 7, since the nodes existing at the boundary have no nodes on one side, the synthesized vector obtained by the above method is used. The size tends to increase toward the inside of the model (that is, the length of the arrow indicating the combined vector tends to increase). On the other hand, the magnitude of the resultant vector is considered to be close to zero as the nodes are present inside. Therefore, when the size of the combined vector is used for determining the boundary node, the boundary node can be efficiently extracted by appropriately setting the threshold value.

  Further, as a method for determining such a threshold value, assuming that the element size (ie, node spacing) shown in FIG. 7 is 1 unit, and the region radius is 2 units, the size of the combined vector of boundary nodes | P | Is 5 units to 5.66 units (edge portion), and the combined vector size | P | of nodes existing within one from the boundary is 2 units to 2.83 units (edge portion). Therefore, when the threshold value is 3 units or more, only the region nodes can be extracted. However, in order to apply to an actual analysis model, the threshold value is set to 2 in order to prevent extraction of region nodes. It is desirable to use more than unit. In practice, a reference element size such as an average size of all element sizes is set, a region radius twice as large as the basic element size is set, and a threshold value of the composite vector size | P | Is preferably set to be twice or more.

(Step S8)
In step S8, the same node shared element search means 17 determines the number of elements sharing the node i (hereinafter referred to as “shared element number”) because the magnitude | P | Count.
(Step S9)
Next, in step S9, the same node shared element search means 17 compares the count number with a preset shared element number threshold value, and the count number is equal to or less than the preset shared element number threshold value ( If YES, the process proceeds to step S10. Otherwise, if the count number exceeds a preset shared element number threshold value (NO), the process proceeds to step S11. However, when counting the number of shared elements, the element table created in step S2 is referred to, and the number of shared elements is determined by searching the element table for elements having node i as a constituent node. And

(Step S10)
In step S10, the same node shared element search means 17 adds and updates the node i to the extraction list, and proceeds to step S11.
(Step S11)
In step S11, the same node shared element search means 17 performs processing so that the processing target of the series of processing in step S3 and subsequent steps is the next node, that is, node (i + 1) whose node number is (i + 1). This process is repeated until the last node of the imported analysis model (that is, the number of nodes: M contacts), and then the finally created extraction list is output by the output means 18. To do.

The threshold for the number of shared elements depends on the element quality of the created analysis model.
FIG. 10 is an explanatory diagram for explaining a node sharing element.
For example, in the case of a two-dimensional element, if the inner angle of the element is created with less than 120 °, in a state where three elements share one node as shown in count = 3 shown in FIG. There is no possibility of being inside. Therefore, the shared element threshold is set to 3. That is, generally, a quotient obtained by dividing 360 ° by the allowable maximum inner angle of the created element is obtained, and a value obtained by subtracting 1 from the minimum natural number equal to or greater than the quotient can be set as the shared element number threshold. Since the two-dimensional element is expanded into three dimensions, the shared element threshold value in this case is set to 6.

  The function of the boundary node extraction system according to the embodiment of the present invention described above will be described using a cubic analysis model shown in FIG. 6. In the prior art, the surface existing at the boundary is shared with the surfaces of other elements. Boundary nodes were extracted focusing on the fact Therefore, select an element, extract the constituent plane of the element, and whether there are other elements that share the constituent nodes of the plane, that is, whether there are other nodes that share the same plane There is a problem in that it is necessary to check whether or not the combination of all the elements, and the process becomes a nested structure and the number of repeated processes increases.

  On the other hand, in the boundary node extraction system according to the embodiment of the present invention, first, as a pre-process, nodes existing on the boundary only by the relative positional relationship of the nodes without using elements (that is, candidates for boundary nodes) ) Is extracted, the number of elements sharing the node is counted, and the boundary node is determined and extracted. When counting the number of shared elements that share this node, the element is referred to, so iterative processing occurs. However, since the number of processes is performed only for the nodes extracted in the preprocessing, Can be reduced.

  More specifically, in the case of the prior art, as described in the background art section, if the flow of FIG. 16 is followed, at least 1798200 iterations have occurred. In the case of such a boundary node extraction system, as shown in the flowchart of FIG. 3, the process is basically completed by a single iterative process for all nodes. However, as described above, repetition occurs in the process of counting the number of elements sharing a node (step S8), and in the case of a hexahedral 8-node element, 8 × 1000 = 8000 repetition processing is performed for one node. It will be. However, the nodes to be subjected to the repetitive processing are also nodes determined by the threshold value in step S7, and it is not necessary to perform the process on all the nodes. If it is assumed that the processing has been performed for all nodes, the number of processing times as many as 1331 × 8000 = 10648000 times is required. Therefore, the number of times of repeated processing can be greatly reduced by determining with a threshold value.

FIG. 4 is another flowchart showing the best processing procedure of the boundary node extraction system according to the embodiment of the present invention.
The process of steps S21 to S27 shown in the figure corresponds to the process of steps S1 to S7 shown in FIG. Steps S21 to S28 are pre-processing, that is, processing for creating an extraction list by extracting nodes (that is, candidates for boundary nodes) existing on the boundary only by the relative positional relationship of the nodes. Steps S29 to S34 are processes for determining the boundary node by counting the number of elements sharing the node from the boundary node candidates extracted in the extraction list. Thus, by dividing the processing steps into the pre-processing and the post-processing, the nested structure in the case of the flowchart in FIG. 3 can be eliminated.

FIG. 5 is another flowchart showing the best processing procedure of the boundary node extraction system according to the embodiment of the present invention.
The process of steps S41 to S42 shown in the figure corresponds to the process of steps S21 to S22 shown in FIG. Moreover, the process of step S43 to S49 shown to the figure respond | corresponds to the process of step S23 to S29 shown in FIG. Furthermore, the processing of steps S50 to S54 shown in the figure corresponds to the processing of steps S30 to S34 shown in FIG. In this way, the processing steps can be divided into pre-processing and post-processing, and both processes can be performed in parallel, thereby improving the processing speed as compared with the flowchart of FIG.

Next, each example according to the embodiment of the present invention will be described with reference to the drawings.
FIG. 8 is a first explanatory diagram showing a display screen as an example of an analysis model in the boundary node extraction system according to the embodiment of the present invention.
FIG. 17 is a second explanatory diagram showing a display screen as an example of an analysis model in the boundary node extraction system according to the embodiment of the present invention.
First, the boundary node extraction system according to the embodiment of the present invention is called from the external storage device 6 (FIG. 1) or the auxiliary storage device 5 (FIG. 1) by the input device 1 and activated. Thereby, for example, analysis model data as shown in FIG. 8 is displayed on the display device 7 (FIG. 1).
By selecting the displayed button of the command unit 31 with the input device 1 or by directly inputting a numerical value into the numerical value display unit of the setting unit 34, the boundary node extraction system executes the above-described functions.
Normally, this boundary node extraction system has a global coordinate system 32 on the display screen shown in FIG. 8, and based on this coordinate system, each of the nodes and elements described above is displayed on the display unit 33 as an analysis model. It is displayed.

The first embodiment will be described below with reference to the flowchart shown in FIG.
First, the analysis model data of the 3D analysis target structure was imported into the boundary node extraction system.
FIG. 17 shows a state in which the imported analysis model is displayed on the display device 7.
The analysis model shown in FIG. 17 has a circular hole, a C, R chamfer, a free-form surface portion, an acute edge, and an obtuse angle edge included in a general mechanical structure. The analysis model data of the two-dimensional analysis target structure shown in FIG. 8 that was created using the basic element size of 0.2 mm in a lump was created by extending in the direction perpendicular to the paper surface of FIG.

  For the called analysis model, the region radius setting means 12 (FIG. 2) sets the region radius, and the threshold setting means 13 sets the threshold value for the size of the combined vector of the position vectors and the threshold value for the number of shared elements. However, as a reference for determining the region radius, a value twice the reference size of the element of the analysis model was set. The analysis model shown in FIG. 17 is created by extending the 2D analysis model data shown in FIG. 8 with a basic element size of 0.2 mm by 0.2 mm, so the basic element size is 0.2 mm. Thus, the area radius was set to 0.4 mm. As the reference value for determining the threshold value of the size of the combined vector, 0.4 mm, which is twice the reference size of the element of the analysis model, was set as in the case of the area radius. The threshold for the number of shared elements depends on the quality of the created element. Since the two-dimensional analysis model shown in FIG. 8 allows an inner angle of 140 °, the threshold value is set to 5 or less in the analysis model of FIG.

Information on the number of elements, the number of nodes, and the element type of the imported analysis model is automatically displayed on the setting unit 34. The above-mentioned area radius and threshold were set by directly inputting to the setting unit 34.
When the execution button of the setting unit 34 is pressed, processing is executed according to the processing procedure of the flowchart shown in FIG.
From the analysis model to be the boundary node extraction target, the node number and the coordinate value of each node with respect to the coordinate system are stored as a node table (step S1), and the element number and the node number constituting the element are stored as an element table. Save (step S2). The storage form of this table is stored in the memory as an array. On the other hand, it is also possible to obtain necessary node information and element information by directly accessing the analysis model data to be imported without holding it in the memory. That is, the analysis model data is created in accordance with a unique format that can be interpreted by an analysis solver that executes analysis calculation, and describes the node information and element information described above.

Next, one node is selected (step S3). Since there are no restrictions on the selection of the nodes, the nodes are selected in order from the smallest node number, and all nodes are repeatedly processed until the following step S11.
Node 1 with node number 1 is selected, and as shown in FIG. 18, nodes existing within the area radius set by area radius setting means 12 (FIG. 2) are extracted (step S4). As a start point, a position vector p having a node in the region as an end point is calculated (step S5). As shown in FIG. 19, in the case of the node 1, since there are 11 nodes in the region, 11 vectors can be created. A composite vector P of these 11 vectors is created, and the magnitude | P | of the composite vector P is calculated (step S6).

In the case of the node 1, the magnitude | P | of the vector P is 2.3044, and it is determined in step S7 that it is larger than the threshold value 0.4. As shown in FIG. 11, when the magnitude of the combined vector is larger than the threshold, it is determined that there is a high possibility that the node exists at the boundary.
Next, the elements sharing node 1 are counted. An element having node 1 as a constituent node is searched from the element table created in step S2, and the number of elements is counted. The component node list of element 1 with element number 1 is called from the element table to check whether node 1 is a component node. This operation is performed for all elements from element 1 to element 27750, and the number of elements having node 1 as a constituent node is counted (step S8). In the middle of this series of processing, in step S9, it is verified whether or not the count number is equal to or less than the common element threshold value. If the count number is equal to or less than the common element threshold value, the node is regarded as a boundary node in step S10. Add to extraction list.

Next, the same processing is performed for the node 2 with the node number 2, and similarly, the boundary nodes are extracted for all the nodes up to the node 31424.
In the course of this series of processing, if there is a composite vector P whose size is smaller than the threshold value, the processing for that node is completed at that time, and the processing shifts to the next node. Similarly, when the count value of the number of elements sharing the node is larger than the threshold value, the node is not regarded as a boundary node, and the process is completed and is not added to the extraction list. FIG. 20 shows a state where the processing of all nodes is completed and the boundary nodes are displayed.

Hereinafter, similarly, the second embodiment will be described using the flowchart shown in FIG. 3 and the two-dimensional analysis model shown in FIG.
First, as in the first example, the analysis model data of the structure to be analyzed was imported into the boundary node extraction system. FIG. 8 shows a state in which the imported analysis model is displayed on the display device 7.
As described above, the two-dimensional analysis model of FIG. 8 has a circular hole, C, R chamfer, free curve portion, acute angle edge, and obtuse angle edge, which are included in a general mechanical structure. Using the automatic element creation algorithm, batch elements were created with a basic element size of 0.2 mm.

  For the called analysis model, the region radius setting means 12 (FIG. 2) sets the region radius, and the threshold setting means 13 sets the threshold value for the size of the combined vector of the position vectors and the threshold value for the number of shared elements. However, as a reference for determining the region radius, a value twice the reference size of the element of the analysis model was set. In the analysis model shown in FIG. 8, since the basic element size is 0.2 mm, the area radius is set to 0.4 mm. As the reference value for determining the threshold value of the size of the combined vector, a value twice the reference size of the element of the analysis model was set, as with the area radius. Since the analysis model shown in FIG. 6 is created with a basic element size of 0.2 mm, the composite vector size threshold is also set to 0.4 mm. The threshold for the number of shared elements depends on the quality of the created element. Since the analysis model shown in FIG. 8 allows an internal angle of 140 °, the threshold value is set to 2 or less according to the threshold value setting method described above.

Information on the number of elements, the number of nodes, and the element type of the imported analysis model is automatically displayed on the setting unit 34. The above-mentioned area radius and threshold were set by directly inputting to the setting unit 34.
When the execution button of the setting unit 34 is pressed, processing is executed according to the processing procedure of the flowchart shown in FIG.
From the analysis model to be the boundary node extraction target, the node number and the coordinate value of each node with respect to the coordinate system are stored as a node table (step S1), and the element number and the node number constituting the element are stored as an element table. Save (step S2). The storage form of this table is stored in the memory as an array. On the other hand, it is also possible to obtain necessary node information and element information by directly accessing the analysis model data to be imported without holding it in the memory. That is, the analysis model data is created in accordance with a unique format that can be interpreted by an analysis solver that executes analysis calculation, and describes the node information and element information described above.

Next, one node is selected (step S3). Since there is no restriction on the selection of the nodes, the nodes are selected in order from the smallest node number, and the process is repeated for all the nodes until the process of step S11 below.
FIG. 12 is an explanatory diagram showing a state where boundary node extraction has been completed for the analysis model shown in FIG. Node 1 with node number 1 is selected, and as shown in FIG. 11, nodes existing within the area radius set by area radius setting means 12 (FIG. 2) are extracted (step S4). As a start point, a position vector p having a node in the region as an end point is calculated (step S5). In the case of the node 1, since there are seven nodes in the region, seven position vectors can be created. A composite vector P of these seven position vectors is created, and the magnitude | P | of the composite vector P is calculated (step S6).

FIG. 11 is an explanatory diagram showing the relationship between the threshold value and the combined vector P of the position vector p whose end point is the node in the region when the boundary node candidate is extracted.
As shown in FIG. 11, the vector P can be drawn, and the magnitude | P | of the vector P is 1.39073, and is determined to be larger than the threshold value (= 0.4) in step S7. As shown in FIG. 11, when the magnitude of the combined vector P is larger than the threshold, it is determined that the node is highly likely to exist at the boundary.
Next, the elements sharing node 1 are counted. The element having node 1 as a constituent node is searched from the element table created in step S2, and the number of elements is counted. The component node list of element 1 with element number 1 is called from the element table to check whether node 1 is a component node. This operation is performed for all elements from element 1 to element 1850, and the number of elements having node 1 as a constituent node is counted (step S8). In the middle of this series of processing, in step S9, it is verified whether or not this count number is less than or equal to the shared element threshold value. If this count number is less than or equal to the shared element threshold value, this node is regarded as a boundary node in step S10. Add to extraction list.

Next, the same processing as described above is performed for the node 2 with the node number 2, and similarly, the boundary nodes are extracted for all the nodes up to the node 1964.
In the course of this series of processing, if there is a composite vector P whose size is smaller than the threshold value, the processing for that node is completed at that time, and the processing for the next node is started. Similarly, when the count value of the number of elements sharing a node is larger than the threshold value, the node is not regarded as a boundary node, and the process is completed and is not added to the extraction list.

FIG. 12 shows a state where the processing of all nodes is completed and the boundary nodes are displayed.
In this embodiment, 2 is adopted as the threshold value of the count number of elements sharing the node, but the threshold value depends on the quality of the created analysis model. For example, when the internal angle of the element is created with less than 120 °, there is no possibility that the case of COUNT = 3 shown in FIG. 10 is an internal node. The analysis model of the present embodiment employs an automatic element dividing means and allows an internal angle of 140 °, so that the boundary node can be extracted by setting the threshold value to 2 or less. FIG. 9 is an explanatory diagram that lists an example of in-region nodes extracted corresponding to each node. That is, all the nodes existing within the area radius set in advance by the area radius setting means 12 with the node i (i = 1 to 1964) as the center are extracted, and the number of each node and each contact number are listed. It is a thing.

The third embodiment will be described below with reference to the flowchart shown in FIG.
In the second embodiment described above, the iterative process (step S8) of searching for and counting elements sharing a node from all elements, the node selection process (step S3) from the initial stage of the flow to the next node selection process (step S11). It has a nested structure that goes into the loop leading up to. However, a node is shared with a series of processes (steps S23 to S29) in which a node is selected, a position vector p is calculated at a node in the region, a boundary node is determined from the magnitude of the combined vector P and listed. A series of processes for searching and counting elements (steps S30 to S34) can be sequentially linked as processes independent of each other.
Therefore, in the third embodiment, the nested structure of the above two processes is eliminated, and the sequential expansion process is performed.

The fourth embodiment will be described below with reference to the flowchart shown in FIG.
In the third embodiment described above, a node is selected, a vector is calculated from the nodes in the region, and a boundary node is determined from the size of the combined vector (steps S23 to S29), and elements sharing the node are included. Processes for searching and counting (Steps S30 to S34 are sequentially connected. In the fourth embodiment, these processes are arranged in parallel and processed independently.

  In this embodiment, a node extraction list is created by the two independent processes executed in parallel, and finally the product of both is taken and narrowed down (step S55), thereby creating a final joint extraction list. Yes. In this embodiment, since the contents of all processes cannot be reflected in the post-processing in order to perform parallel processing, both of the two processes need to be performed on all nodes. Can be processed completely in parallel, so the calculation time can be expected to be shortened.

Next, a fifth embodiment will be described with reference to FIGS.
In the fifth embodiment, a case where a plurality of parts are assembled will be described.
FIG. 12 shows a state in which extraction of boundary nodes has been completed with respect to the analysis model shown in FIG. 8 referred to in each of the foregoing embodiments. In the following description, the analysis model shown in FIG. And the newly set analysis model shown in FIG.
As pre-processing for assembling Parts A and B, boundary nodes were extracted for each part by the processing procedure of any of the above-described embodiments. This process can be executed individually for each part, and is not necessarily synchronized with the time of assembly, and can be performed at any time as long as it is after the completion of model creation.

FIG. 14 is an explanatory diagram showing a state in which a plurality of analysis models are assembled.
The figure shows the state where Part A and Part B are assembled and the extracted boundary nodes are clearly shown. When determining whether or not to consider the contact between Part A and Part B, it is determined that the contact needs to be considered when the value of the shortest distance between both parts is equal to or less than the threshold for determining contact.
Select each node extracted as a boundary from each Part, calculate the distance between the nodes and compare it with the contact determination threshold. If the distance between the nodes is less than or equal to the threshold, A pair of nodes was memorized.

  In this example, 78 nodes were extracted as boundary nodes in Part A, and 229 were extracted in Part B. Next, one node extracted as a boundary node in Part A was selected, and the distance from all the nodes extracted as boundary nodes in Part B was calculated. When the distance is equal to or smaller than the contact determination threshold, the node numbers of both nodes and the distance between the nodes are stored. FIG. 15 shows the results obtained when the contact determination threshold was set to 0.3 mm in this example.

FIG. 15 is an explanatory diagram showing a result of analyzing the distance between nodes of a plurality of analysis models. The figure shows the numbers of two nodes at a distance equal to or smaller than the contact determination threshold and the distance between the nodes. Based on these pieces of information, a contact element can be created separately, or contact can be considered by adding a contact consideration attribute to an element having a node as a constituent element.
In this example, the distance calculation necessary for the distance exploration was performed 78 × 229 = 17862 times. In the case of the prior art, since distance exploration is carried out for all nodes, it will be carried out 307 × 1964 = 602948 times, but it was found that the calculation time can be greatly reduced compared to this. In the above formula, the numerical value 1964 is the total number of nodes in Part A, and the numerical value 307 is the total number of nodes in Part B.

  Note that at least a part of the processing of each component of the boundary node extraction system according to the present invention is executed by computer control, and the above processing is performed by the computer according to the procedure shown in the flowcharts of FIGS. The program to be executed may be distributed after being stored in a computer-readable recording medium such as a semiconductor memory, a CD-ROM, or a magnetic tape. A computer including at least a microcomputer, a personal computer, and a general-purpose computer may read the program from the recording medium and execute the program.

  According to the present invention, when a finite element method is used to verify the design of any structure such as an electronic device, an automobile, and a building, a node existing at the boundary is efficiently obtained from the analysis model of the finite element method. Therefore, it can be suitably applied to tasks such as contact determination, contact element creation, and boundary condition assignment when the analysis target is composed of a plurality of parts.

DESCRIPTION OF SYMBOLS 1 Input device 2 Data processing device 3 Central processing device 4 Main storage device 5 Auxiliary storage device 6 External storage device 7 Display device 8 Output device 11 Input means 12 Area radius setting means 13 Threshold setting means 14 Intra-region node extraction means 15 Vector calculation Means 16 Boundary node determination means 17 Same-node shared element search means 18 Output means

Claims (10)

A boundary node extraction system that extracts nodes existing at the boundary of an analysis model used in numerical analysis,
Based on the position information of the nodes constituting the analysis model, for each of the nodes, the nodes in the region that extract other nodes existing in a circle centered on the node and having a predetermined region radius as a radius Extraction means;
Position vector constructing means for creating a position vector having the node at the center of the circle as the tail and the extracted other node as a tip;
A combined vector calculating means for calculating a combined vector of the position vectors and calculating the magnitude of the vector;
Boundary node extraction means for extracting the node as the center of the circle as a candidate for a boundary node when the size of the combined vector exceeds a predetermined first threshold;
Boundary node extraction system characterized by comprising A shared element counting means for searching for an element of the analysis model sharing the node as the center of the circle based on information of a node constituting the element and counting the number of the elements detected in the search; ,
Boundary node determination means for excluding the node as the center of the circle from the extracted boundary node candidates when the counted number of elements exceeds a predetermined second threshold. The boundary node extraction system according to claim 1, wherein:   The intra-regional node extraction means, the position vector construction means, the combined vector calculation means, the boundary node extraction means, the shared element counting means, and the boundary node determination means execute sequential processing. The boundary node extraction system according to claim 2.   The intra-regional node extracting means, the position vector constituting means, the combined vector calculating means, the boundary node extracting means, the shared element counting means, and the boundary node determining means execute parallel processing. The boundary node extraction system according to claim 2.   When setting a threshold for determining the region radius and the size of the combined vector, an average value of element sizes is obtained based on the data of the analysis model, and the average value is set as a basic element size, and the basic 2. The boundary node extraction system according to claim 1, wherein an integer multiple of an element size is set as a threshold for determining the area radius and the size of the composite vector.   When setting the second threshold, the minimum number of elements that can be shared by internal nodes is calculated based on the maximum inner angle of the elements, and the calculated number of elements is set as the second threshold. The boundary node extraction system according to claim 2. A boundary node extraction method for extracting nodes existing at the boundary of an analysis model used in numerical analysis,
Based on the position information of the nodes constituting the analysis model, for each of the nodes, the nodes in the region that extract other nodes existing in a circle centered on the node and having a predetermined region radius as a radius An extraction step;
A position vector configuration step of creating a position vector having the node at the center of the circle as the tail and the extracted other node as a tip;
Determining a combined vector of the position vectors, and calculating a combined vector calculating step of calculating a magnitude of the vector;
A boundary node extracting step of extracting the node as the center of the circle as a candidate for a boundary node when the size of the combined vector exceeds a predetermined first threshold;
A boundary node extraction method characterized by comprising: A shared element counting step of searching for an element of the analysis model sharing the node as the center of the circle based on information of the node constituting the element and counting the number of the elements detected in the search; ,
A boundary node determination step of excluding the node as the center of the circle from the extracted boundary node candidates when the counted number of elements exceeds a predetermined second threshold; The boundary node extraction method according to claim 7, wherein: A computer that extracts nodes at the boundary of the analysis model used in numerical analysis,
Based on the position information of the nodes constituting the analysis model, for each of the nodes, the nodes in the region that extract other nodes existing in a circle centered on the node and having a predetermined region radius as a radius An extraction step;
A position vector configuration step of creating a position vector having the node at the center of the circle as the tail and the extracted other node as a tip;
Determining a combined vector of the position vectors, and calculating a combined vector calculating step of calculating a magnitude of the vector;
A boundary node extracting step of extracting the node as the center of the circle as a candidate for a boundary node when the size of the combined vector exceeds a predetermined first threshold;
A program for running A shared element counting step of searching for an element of the analysis model sharing the node as the center of the circle based on information of the node constituting the element and counting the number of the elements detected in the search; ,
A boundary node determination step of excluding the node as the center of the circle from the extracted boundary node candidates if the counted number of elements exceeds a predetermined second threshold;
10. The program according to claim 9, further comprising:

Images