JP2800709B2 - Method of segmenting FEM analysis model - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for dividing a region of an FEM analysis model, and more particularly to an FE having a hole inside.
The present invention relates to a method for dividing a region of an FEM analysis model suitable for creating an M (finite element method) analysis model.

[0002]

2. Description of the Related Art In order to secure safety, reduce costs, shorten development periods, and the like, it is necessary to perform structural calculations when designing various structures and structural parts. However, when it has a complicated shape or uses a composite material, structural calculation is very difficult.

Therefore, with the development of computers, the finite element method (FF) capable of discretizing differential equations induced by analytical mechanics and performing approximate high-speed structural calculations.
EM) has been used.

[0004] In the FEM, the operation time and the operation accuracy greatly differ depending on the way of dividing the target shape into elements. for that reason,
Performing calculations using FEM requires many years of experience,
A specialized operator has performed analysis processing at the request of the designer.

In this case, it takes time until an effective result is obtained, and there is a problem that it is not possible to shorten the design period, which is an initial purpose, and there is a problem that the use of the FEM is limited.

Therefore, an automatic mesh creation technique has been introduced so that a designer himself can perform FEM analysis.

[0007]

However, in the above conventional example, as shown in FIG. 4, in the case of a shape having a hole inside, mesh cannot be automatically generated as it is, Since it is manually divided into several rectangles consisting of the outlines that form the shape and the interior lines that form the holes, and the automatic mesh generation for these rectangles, labor and time are reduced. In addition, there is an inconvenience that the analysis result is different due to the individual difference of the rectangular division method.

[0008]

SUMMARY OF THE INVENTION An object of the present invention is to improve the disadvantages of the prior art. In particular, even when the target shape has a hole inside, the optimum mesh division is automatically performed in the same manner as when there is no hole. It is an object of the present invention to provide a method for dividing an FEM analysis model that can be performed.

[0009]

Therefore SUMMARY OF THE INVENTION In the present invention, co
Object of the finite element method for performing structural calculations using a computer
Area of FEM analysis model that automatically divides shape area
A division method, in which lines and holes that form the outer shape of the target shape are
The line to be formed is extracted and the target shape is
Divided into regions having lines forming holes and lines forming holes
Line segment dividing step, and the line segment dividing step
Shape recognition process for recognizing the shape of each region
The shape of each area recognized in the recognition process is
And a registration step of registering as a data. Furthermore, the line
In the dividing step, the line forming the outer shape and the line of the hole are reversed.
The process of setting the vector and the outline from the top of the hole line
Drop the perpendicular to the line to be formed, and
The process of registering the intersection with the line of
From a point at the tip of the line or hole
Search for nodes, and if the node is at the intersection with the perpendicular
Search for the opposing node on the perpendicular
It has a configuration that has a process.

[0010]

[0011]

In the present invention, the above-mentioned object is achieved.

[0013]

In the line segment dividing step, a perpendicular line is dropped from the intersection, that is, the vertex, of each line forming the hole to the line forming the external shape of the target shape, thereby forming the line forming the external shape of the target shape. The target shape is divided into an area including one line forming a hole.

In the shape recognition step, the area created in the line segment dividing step is searched from a certain vertex in a predetermined direction, and the shape is recognized by finding each vertex position of the area. This recognition processing is performed on all the regions created in the line segment dividing step.

In the registration step, the shape recognition data in the shape recognition step is registered. Automatic mesh creation processing is subsequently performed based on the shape recognition data.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to FIGS.

The embodiment shown in FIG. 1 includes a line segment dividing step 1 for dividing a target shape into an area including a line segment forming an outer shape of the target shape and a line segment forming a hole; Shape recognition step 2 for recognizing the shape of each area divided in
And a registration step 3 for registering the shape of each area recognized in the shape recognition step 2.

Next, the operation of this embodiment will be described with reference to the flowchart of FIG. Here, for the sake of simplicity, the line segment ab and the line segment bc as shown in FIG.
In the case of a shape in which a rectangular hole surrounded by a line segment ef, a line segment fg, a line segment gh, and a line segment he exists inside a rectangle surrounded by the line segment cd and the line segment da, explain.

In the following description, a line forming the outer shape of the target shape is called an outer shape line, a hole shape is called an inner shape, and a line forming a hole is called an inner shape line.

Further, the vertices of the outer shape and the inner shape are called nodes.

(1). In the line segment dividing step 1, mutually opposite vectors are assigned to the outer shape line and the inner shape line of the target shape (step S1 in FIG. 2). That is, if the outer shape is clockwise, the inner shape is counterclockwise. Naturally, if the outer shape is counterclockwise, the inner shape is clockwise.

Here, as shown in FIG.
The outer shape abcd is clockwise, and the inner shape efgh is counterclockwise.

(2). Further, in the line segment dividing step 1, a perpendicular line is drawn from each inner node to the closest external line segment, and the line segment is divided at the intersection (step S in FIG. 2).
2).

Here, as shown in FIG.
A perpendicular line is drawn from the inner node e to the outer line segment ab, and the intersection is defined as a point i. Similarly, a perpendicular line is drawn from the inner node h to the outer line segment bc, and its intersection is set to a point j. A perpendicular line is drawn from the inner node g to the outer line segment cd, and the intersection point is set to a point k.
Then, a perpendicular line is drawn from the inner node f to the outer line segment da, and the intersection is defined as a point ι.

Then, data of each perpendicular line (hereinafter, referred to as a dividing line) (for example, coordinate data of a starting point and an ending point) and coordinate data of each intersection (hereinafter, referred to as a dividing point) are stored.

(3). When the division in the line segment dividing step 1 is completed, the process proceeds to the shape recognition step 2. In the shape recognition step 2, first, a division start point of each area is designated (step S3 in FIG. 2).

Here, as shown in FIG.
Points a, b, c, and d are division start points.

Then, the number of division start points and their coordinate data are stored.

(4). Next, in the shape recognition step 2, starting from one of the division start points defined in the above process (3), a division point is searched in the vector direction of the outer shape to obtain position data of the division point (step S4 in FIG. 2). ).

Here, as shown in FIG.
The division point is searched in the vector direction, that is, clockwise from the point a which is the first start point, and the position data of the division point i is recognized.

(5). Further, in the shape recognition step 2, an inner shape is formed along the division line from the division point, and the next node position data is obtained (step S5 in FIG. 2).

Here, as shown in FIG.
Proceeding clockwise along the dividing line ie from the dividing point i, the node e
Recognize the position data.

(6). Then, a dividing line is searched in the vector direction of the closed contour loop, and the next node position data is obtained (step S6 in FIG. 2).

Here, as shown in FIG.
Proceeding clockwise from node e, the position data of node f is recognized.

(7). Subsequently, in the shape recognition step 2, the process returns to the outer shape along the division line, and the next division point position data is obtained (step S7 in FIG. 2).

Here, as shown in FIG.
Proceed clockwise from the node f along the dividing line fι to the dividing point
Recognize the position data.

(8). Then, it is checked whether or not the position is a start point (step S8 in FIG. 2).

If it is not the start point, the vector direction of the external closed loop is further searched to obtain the next node position data (step S9 in FIG. 2).

On the other hand, if it is the starting point, the registration step 3
Then, the area is registered as a closed loop (step S10 in FIG. 2).

Here, as shown in FIG.
Since the division point ι is not the start point, the search is again performed clockwise from the division point ι to recognize the node position data of the node a.

Then, since the node a is the starting point, the process shifts to the registration step 3, and as shown in FIG. 3 (g), the line segment ai, the line segment ie, the line segment ef, the line segment fι, and the line segment The area surrounded by ia is defined as one closed loop, and the area data is registered.

(9). When the registration processing in the registration step 3 is completed, the process proceeds to the shape recognition step 2 again. Shape recognition process 2
Then, the next unregistered start point is searched in the vector direction of the contour closed loop (step S11 in FIG. 2).

If there is no unregistered start point (step S12 in FIG. 2), the process ends.

On the other hand, if there is a start point that has not been registered (step S12 in FIG. 2), a closed loop is detected in the same manner as above, using that point as a new start point, and the area data of the closed loop is registered. Register by.

Here, as shown in FIG.
Since the node b recognized by searching the outer shape from the node a clockwise is not registered, a closed loop starting from the node b is detected, and the line segment bj, the line segment jh, the line segment he, and the line segment are detected. The area data of the closed loop surrounded by ei and the line segment ib is registered.

Further, similarly, a closed loop starting from the nodes c and d is detected, and the line segment ck and the line segment k are detected.
Data of an area surrounded by g, a line gh, a line hj, and a line jc, and data of an area surrounded by a line dι, a line lf, a line fg, a line gk, and a line kd Register

(10). When the data registration of the closed loop is completed for all the start points, the process shifts to the conventional automatic mesh generation processing (step S13 in FIG. 2).

By the above-described processes, even when the target shape has a hole, it can be automatically divided into a combination of optimal shapes without holes, and even an inexperienced designer can easily and optimally obtain an optimal FEM in a short time. You can create an analysis model.

[0049]

Since the present invention is constructed and functions as described above, according to the present invention , even if the target shape has a hole,
Set the line of the hole and the line of the outline to the opposite vector,
Set a perpendicular line between them, and search for nodes in the vector direction.
Search and divide into multiple regions before generating mesh
The computer automatically sets the optimal shape without holes
This makes it possible to create a FEM analysis model in a short time and easily even for a beginner, and there is no difference in analysis results due to the difference in the division method, thereby improving reliability. Excellent FEM
A method for segmenting an analysis model can be provided.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing one embodiment of the present invention.

FIG. 2 is a flowchart for explaining the operation of the present invention.

FIG. 3 is an explanatory diagram for explaining the operation of the present invention;
3A is a diagram corresponding to S1 in FIG. 2, FIG. 3B is a diagram corresponding to S2 in FIG. 2, FIG. 3C is a diagram corresponding to S4 in FIG. 3D is a diagram corresponding to S5 of FIG. 2, and FIG. 3E is a diagram corresponding to S6 of FIG.
FIG. 3F is a diagram corresponding to S9 in FIG. 2, and FIG.
3 (h) is a diagram corresponding to S13 of FIG.

FIG. 4 is a flowchart for explaining the operation of the conventional example.

[Explanation of symbols]

 1 Line segmenting process 2 Shape recognition process 3 Registration process

Continuation of the front page (58) Field surveyed (Int.Cl. ⁶ , DB name) G06T 17/20 JICST file (JOIS)

Claims (1)

(57) [Claims] (1)Perform structural calculations using a computer
FEM that automatically divides the area of the target shape of the finite element method
An area division method for an analysis model, Lines forming the outer shape of the target shape and lines forming the holes
Extracting the target shape and the line forming the outer shape
Line segmentation step of dividing into regions having lines forming holes
And the shape of each area divided in this line segment dividing step
Shape recognition process for recognizing
Register the shape of each recognized area as shape recognition data
Registration process, The line segment dividing step includes the step of forming a line between the line forming the outer shape and the hole.
Setting the line and the vector in the opposite direction; and
And drop a perpendicular from the vertex to the line forming the outline
The intersection between the perpendicular and the line of the target shape is registered as a node.
Recording process, The shape recognizing step may be a line or the hole forming the outer shape.
A node in the vector direction from a point at the end of
If the process is at the intersection with the perpendicular,
Searching for opposing nodes on the perpendicular.
That A method for dividing a region of an FEM analysis model as a feature.