JPH0863619A - Device and method for automatically preparing part - Google Patents

Abstract

PURPOSE: To provide automatical part preparing device and method capable of recognizing an analytical featured shape part from a shape to be analyzed, uniformly dividing the analytical featured shape part by selecting pattern data on the recognized part and executing adjustment between plural patterns and the automatic division of the outside of the patterns. CONSTITUTION: Proper part shape data are prepared from shape data read out of a shape data memory 1 by the use of respective means in an arithmetic unit 5, condition data stored in a shape data memory 1, knowledge data stored in a knowledge data memory 3, and pattern data stored in a pattern data memory 2, the shape is displayed on a display device 6 and analytical data are outputted to an output device 7. Consequently optimum and uniform parts matched with an analytical purpose and independent of an analyzer can be automatically prepared.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to data generation of an analysis code of the finite element method, and in particular, the analysis target shape is automatically divided into parts which are aggregates of meshes (elements) and the part shape data is displayed. And a part automatic generation apparatus and a generation method for outputting.

[0002]

2. Description of the Related Art A conventional part automatic generation apparatus is disclosed in Japanese Patent Laid-Open No.
As described in JP-A-120385, a method of dividing a part by adding a line segment to the analysis target shape from the beginning in dividing the part has been adopted.

[0003]

However, this method has a problem that the line segment is arbitrarily added, so that the division of the analytical characteristic shape portion, which is important in the evaluation of the analysis result, is not uniform.

An object of the present invention is to provide an automatic part generation apparatus and a method for uniformizing division of analysis feature shape parts of shape data, and enabling adjustment between patterns and automatic division outside the pattern. is there.

[0005]

In order to achieve the above object, the present invention is an apparatus for generating part shape data of a finite element method analysis code, and a shape for storing shape data to be analyzed and condition data. A data memory, a pattern data memory in which pattern data is stored in advance, and a knowledge data memory in which empirical knowledge and engineering knowledge of an analyst are stored in advance, and the data is arithmetically processed to automatically generate the part shape data. In the part automatic generation device having a calculation device for performing the following, the calculation device stores shape data input / output means for reading the shape data and condition data previously stored in the shape data memory, and the pattern data in the pattern data memory. Pattern data input means for reading and knowledge for reading the knowledge data in the knowledge data memory Data pattern input means, characteristic shape part calculation means for calculating an analytical characteristic shape part from the shape data and the knowledge data, and optimum pattern data from the pattern data by the analytical characteristic shape part, the condition data and the knowledge data. Pattern selecting means for automatically selecting different patterns, pattern pasting means for pasting the optimum pattern on the shape data, pattern overlapping portion processing means for processing an overlapping portion between the pasted patterns, and the pasting There is provided a part automatic generation device comprising a pattern external processing means for processing the pattern outside of a region other than the attached pattern.

Another means for achieving the above object is a method for generating part shape data of a finite element method analysis code, which is a shape data memory for storing shape data and condition data to be analyzed, and pattern data. In a pattern data memory in which is previously stored, and the respective data read out from the knowledge data memory in which the empirical knowledge and engineering knowledge of the analyst are stored in advance, in a part automatic generation method for automatically generating the part shape data. , The arithmetic processing reads the shape data and condition data stored in advance in the shape data memory, the pattern data in the pattern data memory, and the respective knowledge data in the knowledge data memory, and reads the shape data and the shape data. An analytical characteristic shape part is calculated from knowledge data, and the analytical characteristic shape part and the condition are calculated. The optimum pattern is automatically selected from the pattern data based on the data and the knowledge data, the optimum pattern is pasted on the shape data, the overlapping portion between the pasted patterns is processed, and the pasted pattern is pasted. There is provided a part automatic generation method characterized by processing the outside of the pattern in an area other than the pattern.

[0007]

According to the present invention, in order to divide an analytical feature shape part into parts, the analysis target shape is an object-oriented data structure in which the type of shape is recognized, and the analyst's empirical and engineering knowledge. It is possible to make the division uniform by making a determination based on the accumulated knowledge data and selecting a matching pattern from the pattern data. In the adjustment between patterns, it is possible to automatically divide the patterns by searching for overlapping portions of the patterns and geometrically adding and deleting line segments. For the division outside the pattern, automatic division can be realized by generating a straight line on the grid from the nodes included in the shape data and making the part recognize the range surrounded by the straight line. In this way, the empirical knowledge of the analyst is converted into data, and the knowledge data with engineering knowledge added to it is used as the criterion.
It is possible to efficiently generate a uniform part that does not depend on the analyst.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A part automatic generation apparatus and method according to an embodiment of the present invention will be described below.

FIG. 1 shows the configuration of an automatic part generation apparatus according to an embodiment of the present invention. A shape data memory 1 for storing shape data and condition data to be analyzed, a pattern data memory 2 for storing pattern data, and a knowledge data memory 3 for storing empirical knowledge and engineering knowledge of an analyst.
An input device 4 for an analyst to generate an instruction to generate part shape data, an arithmetic device 5 for performing various arithmetic processes for generation, a display device 6 for displaying the generated part shape data, and , And an output device 7 for outputting as analysis data.

The arithmetic unit 5 includes shape data input / output means 501 for reading shape data and condition data from the shape data memory 1, pattern data input means 502 for fetching the optimum pattern shape from the pattern data memory 2, and knowledge data memory 3. Knowledge data input means 503 for taking in knowledge data relating to analytical feature part calculation; and feature shape part calculation means 504 for calculating an analytical feature shape part from shape data from the shape data memory 1 and knowledge data from the knowledge data memory 3. Pattern selecting means 505 for automatically selecting an optimum pattern from the pattern data in the pattern data memory 2 based on the analytical characteristic shape part, condition data and knowledge data, and pasting the automatically selected optimum pattern shape on the read shape data. Pattern pasting means 506 to be pasted and pasting The pattern overlapping portion processing means 507 for processing the overlapping portion of the processed patterns, the pattern external processing means 508 for processing the area other than the pasted patterns, and the processed and generated part shape data are displayed on the display device 6. The display processing means 509 for performing the processing and the analysis data output means 510 for outputting the analysis data of the generated part shape data to the output device 7. The pattern overlapping portion processing means 507 calculates the overlapping portion between the pasted patterns, and the pattern overlapping portion calculating means 507.
2 and a pattern overlapping portion adjusting means 5074 for adjusting the calculated pattern overlapping portion. The pattern external processing means 508 includes a pattern external calculating means 5082 for calculating the pattern external and a pattern external adjusting means 5084 for adjusting the calculated pattern external.

FIG. 2 shows a flowchart of the processing procedure of the arithmetic unit 5 of FIG. First, in step 8, input device 4
From the shape data memory 1, the shape data is read from the shape data memory 1 into the shape data input / output means 501 of the arithmetic unit 5. The shape data is composed of faces, line segments, and points. A face is a group of line segments, and a line segment is a group of points. FIG. 3 shows an example of the configuration of the shape data. FIG. 4 shows an example of the shape data 100 obtained by displaying the shape data read in step 8 on the display device 6 through the display processing means 509.

Next, in step 9, the analytical feature shape portion 200 of the shape data read in step 8 is calculated by the feature shape portion calculation means 504. Here, the analytical characteristic shape portion 200 includes a surface surrounded by the contour of the shape data, a hole included in the contour, and an uneven portion on the contour as shown in FIG. It is defined as In the calculation of the analytical characteristic shape portion 200, first, the point sequence forming the surface of the shape data read in step 8 is loaded into the pattern pasting means 506. Next, the knowledge data regarding the calculation of the analytic characteristic shape portion 200 is loaded from the knowledge data memory 3 into the knowledge data input means 503. Next, the point sequence data of the shape data fetched by the pattern pasting means 506 and the analytical characteristic shape portion 200 fetched by the knowledge data input means 503.
The presence / absence of the analytical characteristic shape portion 200 is sequentially determined by the characteristic shape portion calculation means 504 from the knowledge data regarding the calculation. For example, if an arbitrary surface is taken out and all the points forming the surface are not included in the inside of the other surface, the surface is defined as the contour portion. For example, if all the points that form a surface are included in the inside of the contour portion, the surface is defined as a hole portion. For example, the continuous arbitrary four points are extracted from the points forming the contour portion, and the first
The center of the circle calculated from the point, the second point, and the third point, and the second
The centers of the circles calculated from the points, the third point, and the fourth point coincide,
In addition, if the center is included in the contour portion, it is a convex portion.
For example, a continuous arbitrary four points are extracted from the points forming the contour portion, the center of the circle calculated from the first point, the second point, and the third point, and the second point, the third point, and the fourth point. If the centers of the circles calculated from are coincident with each other and the center is included outside the contour portion, it is determined as a concave portion. FIG. 5 shows an example of the knowledge data 300 regarding the calculation of the analytical feature shape portion 200.

Next, in step 10, an optimum pattern is selected for the analytical characteristic shape portion 200 calculated in step 9. First, the condition data necessary for pattern selection from the shape data memory 1 is loaded into the shape data input / output unit 501, and the knowledge data regarding pattern selection from the knowledge data memory 3 is loaded into the knowledge data input unit 503, and the loaded condition data and knowledge are read. The data is analyzed by the pattern selecting means 505 in step 9 and the analytical feature shape portion 20 is calculated.
The optimum pattern is determined by comparing with 0, and the optimum pattern shape is loaded from the pattern data memory 2 into the pattern data input means 502. FIG. 6 shows an example of the optimum pattern taken into the pattern data input section. (A) shows an example of the optimal pattern of a hole part, (b) shows an example of the optimal pattern of a concave part, (c) shows an example of the optimal pattern of a convex part.

In the next step 11, the optimum pattern selected in step 10 is pasted on the shape data by the pattern pasting means 506. This paste is step 10
This can be achieved by adding the shape data of the optimum pattern selected in step 8 to the shape data acquired in step 8. FIG. 7 shows an example in which the shape data to which the optimum pattern is attached is displayed on the display device 6 through the display processing means 510.

Next, in step 12, the pattern overlap portion calculation means 5072 calculates the overlap portion between the patterns.

FIG. 8 shows a flow chart of the pattern overlap portion calculation procedure. First, in step 18, one arbitrary analytical feature shape portion 200 is extracted. Then step 1
At 9, any other arbitrary analytical feature shape 200 is extracted. Next, in step 20, the points forming the analytical feature shape portion 200 extracted in step 19 are extracted one by one, and the analytical feature shape portion 200 extracted in step 18 is extracted.
The inside / outside determination of whether it exists inside or outside is performed. In step 21, if it is determined in the inside / outside determination of the point extracted in step 20 that even one point is inside, the process proceeds to step 22 and the analytical feature shape part 200 extracted in step 18 and the analysis extracted in step 19. That the characteristic feature shape portions 200 overlap each other. The above processing from step 18 to step 22 is performed for all the analytical feature shape portions 200.

Next, in step 13, the pattern overlapping portion adjusting means 5074 adjusts the pattern overlapping portion calculated in step 12. Hereinafter, an example of the procedure of the adjustment processing of the pattern overlapping portion will be described with reference to FIGS. 9 and 10. 9 and 10 show an example of adjustment processing of the overlapping portion of the hole and the recess. In FIG. 9, the pattern of holes is referred to as pattern a and the pattern of recesses is referred to as pattern b. In the flowchart of FIG. 10, first, at step 23, an intersection ab between the straight line 1 passing through the centers of the patterns a and b and the patterns a and b.
An intersection A with a straight line connecting 1 and ab2 is calculated, and then a straight line m orthogonal to the straight line 1 and passing through the point A is defined. Next, at step 24, intersections B, C, B ', C'of the straight line m and the outlines of the patterns a, b are calculated. Next step 25
Then, of the four points calculated in step 24, points B and C far from the point A are defined. Next, at step 26, point B,
Straight lines nb 'and nc' which are parallel to the outlines nb and nc of the pattern facing C and pass through points B and C are defined. Next, at step 27, points b1, b2, a1 on the outlines nb, nc,
Points b1 ', a1', a obtained by translating points b1, a1, a2 outside the overlapping portion of a2, a3 on straight lines nb ', nc'
Calculate 2 '. Next, for pattern a,
In 1 ′, a2 is moved to a2 ′, the straight line a2a3 and the straight line a3a4 are deleted, and the straight line a2′C, the straight line CB, and the straight line Ba4 are inserted. Next, for pattern b, b1 is moved to b1 ′, straight lines b1b2 and straight lines b2b3 are deleted, and straight lines b1′B, straight lines BC, and straight lines Cb3 are inserted. In the same manner, adjustments are made for all overlapping parts. FIG. 11 shows the display processing means 50 with all the overlapping parts adjusted.
9 shows an example of displaying on the display device 6 through 9.

Next, in step 14, the pattern outside calculation means 5082 calculates the outside of the pattern. Here, the outside of the pattern means an area other than the portion where the pattern is pasted and adjusted inside the contour portion.

FIG. 12 shows a flow chart of the pattern external calculation processing. First, in step 28, an arbitrary one of the patterns adjusted in step 15 of FIG. 2 is taken out, and an arbitrary point constituting the pattern is taken out. Next, in step 29, it is judged whether or not the extracted one point is included in another pattern. If it is not included, the process proceeds to step 30, and if it is included, the process returns to step 28 to extract the next point. In step 30, regarding the point determined not to be included in step 29, one line segment passing through the point and in the coordinate base axis direction is calculated. Here, for example, one coordinate base axis direction may be the X axis. Next step 3
In step 1, it is determined whether or not the line segment calculated in step 30 is included in the own pattern. If it is included, the process returns to step 28 to extract the next point. If not included, the intersections of all other patterns and the contour portion are calculated. Next, in step 33, the first intersecting point among the intersections calculated in step 23 and the point extracted in step 28 are connected. Hereinafter, the processing from step 28 to step 33 is performed for all the points forming the pattern, that is, all the nodes.

Next, in step 15, the pattern outside adjustment means 5084 adjusts the outside of the pattern calculated in step 14. First, the knowledge data input means 50 is provided with knowledge data regarding the pattern external adjustment from the knowledge data memory 3.
Take in 3. Next, the outside of the pattern is adjusted while checking the data with the outside of the pattern calculated in step 14. For example, with respect to two line segments that are parallel to each other and have opposite directions, the line segment with the longer length when the line distance is d or less is deleted, and the end point of the shorter line segment is replaced with the start point of the long line segment. . Here, d can be set to 5 mm, for example. For example, if two line segments that are parallel to each other and have opposite directions are formed, and the angle between the line segment connecting the start points and one of the two line segments is θ.
Delete the line segment with the longer time length below and replace the end point of the shorter line segment with the start point of the long line segment. Here, θ can be set to, for example, 20 degrees. Further, for example, when the adjusted line segment intersects the pattern, the adjustment process is canceled. In other words, a straight line is generated on the grid from the nodes of the part shape data, the range surrounded by the straight line is defined as a part, and the range determined to be a narrow part is automatically adjusted based on the knowledge information in the knowledge data memory 3. FIG. 13 shows an example of the knowledge data 310 regarding the pattern external adjustment.

Next, in step 16, the part shape data adjusted and processed outside the pattern in step 15 is displayed on the display device 6 through the display processing means 509. 14
An example in which the part shape data 600 subjected to the processing from step 8 to step 15 is displayed on the display device 6 is shown in FIG.

In step 17, the analysis data of the part shape data 600 processed in steps 8 to 15 is passed through the analysis data output means 510 to the output device 7.
It can be output to.

[0023]

According to the present invention, an empirical knowledge and an engineering knowledge of an analyst are used as criteria to automatically generate an optimum part that does not depend on the analyst and that is uniform with the purpose of analysis. it can. Further, since the parts can be automatically divided based on the pattern data or the like, the interactive processing of the analyst can be reduced and the working time can be reduced.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a part automatic generation device according to an embodiment of the present invention.

FIG. 2 is a flowchart of a processing procedure of the arithmetic unit 5 of FIG.

FIG. 3 is an explanatory diagram showing a configuration of shape data.

FIG. 4 is an explanatory diagram showing an example of shape data.

FIG. 5 is an explanatory diagram showing an example of knowledge data for analysis feature portion calculation.

FIG. 6 is a diagram showing an example of an optimum pattern.

FIG. 7 is a diagram showing an example of shape data to which an optimum pattern is attached.

FIG. 8 is a flowchart of a pattern overlapping portion calculation processing procedure.

FIG. 9 is a diagram illustrating an example of pattern overlapping portion adjustment.

FIG. 10 is a flowchart of a pattern overlapping portion adjustment processing procedure.

FIG. 11 is a diagram showing an example of shape data after adjustment of a pattern overlapping portion.

FIG. 12 is a flowchart of a pattern external calculation processing procedure.

FIG. 13 is an explanatory diagram showing an example of knowledge data of external pattern adjustment.

FIG. 14 is a diagram showing an example of automatically generated part shape data.

[Explanation of symbols]

1 ... Shape data memory, 2 ... Pattern data memory, 3
… Knowledge data memory, 4… Input device, 5… Arithmetic device, 5
01 ... Shape data input / output means, 502 ... Pattern data input means, 503 ... Knowledge data input means, 504 ... Feature shape part calculation means, 505 ... Pattern selection means, 506 ...
Pattern pasting means, 507 ... pattern overlapping portion processing means, 5072 ... pattern overlapping portion calculating means, 5074
... pattern overlapping part adjusting means, 508 ... pattern external processing means, 5082 ... pattern external calculating means, 5084 ...
Pattern external adjustment means, 509 ... display processing means, 510
... Analysis data output means, 6 ... Display device, 7 ... Output device

Front Page Continuation (72) Inventor Eiji Suzuki 3-2-1 Sachimachi, Hitachi City, Ibaraki Prefecture Hitachi Engineering Co., Ltd. (72) Inventor Taiichi Kamieno 3-2-1 Sachimachi, Hitachi City, Ibaraki Prefecture In Hitachi Engineering Co., Ltd. (72) Inventor Takashi Nakao 32-1, Sachimachi, Hitachi City, Ibaraki Prefecture In Hitachi Engineering Co., Ltd.

Claims (7)

[Claims] 1. A device for generating part shape data of a finite element method analysis code, comprising a shape data memory for storing analysis target shape data and condition data, a pattern data memory for storing pattern data in advance, and an analysis. In a part automatic generation device having a knowledge data memory that stores in advance empirical knowledge and engineering knowledge of a person, and having a calculation device that performs arithmetic processing on the data in order to automatically generate the part shape data, Shape data input / output means for reading the shape data and condition data stored in advance in the shape data memory, pattern data input means for reading the pattern data in the pattern data memory, and knowledge data in the knowledge data memory. From the knowledge data input means to read, the shape data and the knowledge data A characteristic shape part calculating means for calculating an analytical characteristic shape part; a pattern selecting means for automatically selecting an optimum pattern from the pattern data based on the analytical characteristic shape part, the condition data and the knowledge data; Pattern pasting means for pasting a pattern onto the shape data, pattern overlapping portion processing means for processing an overlapping portion of the pasted patterns, and a pattern for processing a pattern outside the area other than the pasted pattern An automatic part generation apparatus comprising external processing means. 2. The pattern overlap portion processing means according to claim 1, wherein the pattern overlap portion calculating means calculates an overlap portion between the patterns, and the pattern overlap portion adjusts the calculated overlap portion by a geometric process. An automatic part generation apparatus comprising a section adjusting means. 3. The pattern external processing means according to claim 1, comprising pattern external calculation means for calculating an area other than the pattern, and pattern external adjustment means for adjusting the area other than the calculated pattern with the knowledge data. An automatic part generation device comprising: 4. A method for generating part shape data of a finite element method analysis code, comprising: a shape data memory for storing shape data and condition data to be analyzed; a pattern data memory for storing pattern data in advance; Person's empirical knowledge and engineering knowledge are stored in advance in a knowledge data memory, and each piece of data read out from the data is arithmetically processed to automatically generate the part shape data. In the part automatic generation method, the arithmetic processing is the shape data memory. The shape data and condition data stored in advance, the pattern data in the pattern data memory, and the respective knowledge data in the knowledge data memory are read out, and an analytical characteristic shape part is calculated from the shape data and the knowledge data. However, the pattern is determined by the analytical characteristic shape portion, the condition data, and the knowledge data. The optimum pattern is automatically selected from the pattern data and the optimum pattern is pasted on the shape data.
An automatic part generation method characterized by processing an overlapping portion of the pasted patterns and processing the outside of the pattern in an area other than the pasted pattern. 5. The processing of the overlapping portion between the patterns according to claim 4, wherein the overlapping portion between the patterns is calculated,
An automatic part generation method, characterized in that the calculated overlapping portion is adjusted by geometric processing. 6. The automatic part generation method according to claim 4, wherein in the process outside the pattern, a region other than the pattern is calculated, and the region other than the calculated pattern is adjusted by the knowledge data. 7. The adjustment of an area other than the pattern according to claim 6, wherein a straight line is generated from a node of the area on a grid and a range surrounded by the straight line is defined as a part, and knowledge information of the knowledge data memory is defined. An automatic part generation method characterized by automatically adjusting the range determined to be a narrow part.