CN110660135A

CN110660135A - Method for realizing wavefront construction by utilizing triangular gridding rays

Info

Publication number: CN110660135A
Application number: CN201910894471.4A
Authority: CN
Inventors: 胡叶正; 沈天晶; 徐云贵; 黄旭日; 曹卫平; 唐静
Original assignee: Southwest Petroleum University
Current assignee: Southwest Petroleum University
Priority date: 2019-09-20
Filing date: 2019-09-20
Publication date: 2020-01-07

Abstract

A method for realizing wavefront construction by utilizing triangular gridding rays is characterized in that initial rays are divided by utilizing triangular gridding based on model data; and then, obtaining and recording the propagation data of each ray through calculation, if the distance between the monitored rays is larger than a certain critical value, adding one ray in the distance between the monitored rays, initializing, and calculating the position where the distance between the propagation data and the position where the distance is too large. The method can make the rays more uniform, reduce blind spots and poles, and has higher accuracy of filling the rays than the traditional method.

Description

Method for realizing wavefront construction by utilizing triangular gridding rays

Technical Field

The invention relates to the technical field of geological survey, in particular to a novel method for realizing wavefront construction.

Background

The wave front construction method is that a kinematics and dynamics ray tracing equation set is solved by adopting a numerical calculation method to calculate ray propagation data, then under the same travel time, the end points of all rays form a wave front, and then the travel time and the amplitude of a target point are interpolated according to the relative positions of the target point, the wave front and the rays. There are two problems in the conventional method: firstly, in the three-dimensional situation, the division of the initial rays is performed by using the traditional grid division, so that the model has two poles, the ray density at the poles is very high, the calculation amount is very large, the ray division is not uniform, and the probability of occurrence of blind spots is increased; secondly, when the ray distance is continuously increased in the ray propagation process, calculation is very inaccurate, the traditional method adopts an interpolation method, and one ray is estimated from two rays to achieve the purpose of filling the ray distance, but the estimation method is not high in accuracy.

Disclosure of Invention

The invention provides a method for realizing wavefront construction by utilizing triangular gridding rays, which comprises the steps of firstly obtaining a speed model, and then gridding initial rays by using a triangular grid; and secondly, calculating ray propagation data, constructing a wavefront according to the sampling time, increasing rays if the ray distance is greater than a certain critical value, and calculating the propagation data of the increased rays from an initial point so as to solve the problems of ray nonuniformity and inaccurate interpolation in a wavefront construction method.

According to one aspect of the invention, a method for triangulating an initial ray is provided, which comprises the following steps:

step S1, obtaining model data, then selecting an excitation point, and taking the excitation point as a sphere center as a sphere as an initial wavefront;

and step S2, dividing the grid for the first time and numbering, wherein the grid for the first time is a regular octahedron and has six nodes in total, so that only six rays are provided.

In step S3, propagation data calculation is performed on the known six rays.

And step S4, detecting whether the ray density meets the requirement, and completing primary wave field construction if the ray density meets the requirement. Otherwise, the mesh is refined once, and for one surface of the octahedron, the octahedron is a regular triangle, and one regular triangle can be divided into four identical small regular triangles. The division is done for each face of the octahedron to obtain the secondary small triangle data.

In step S5, since the coordinates of the vertices of the divided small triangle data are corrected, that is, the divided triangle is still inside the sphere, all the nodes of the small triangle are extracted to the spherical surface by the distance formula. And the nodes are labeled in a grading way, and a tree-shaped data structure is created for storage and utilization.

Step S6, repeat step S4 and step S5 until the ray density reaches the standard. And completing the construction of the primary wavefront.

And step S7, along with the ray propagation and the wave front construction, if the distance between the rays is too large to meet the requirement, adding one ray between the two rays, and calculating propagation data from the origin.

On the basis of the scheme, the distance formula is as follows:

fine division for the first time, distance from a node to a spherical surface: d ═ vt (1-cos45 °)

Fine division for the second time, wherein the distance from the node to the spherical surface is as follows: d ═ vt (1-vtcos 22.5 °)

……

Fine division for the nth time, wherein the distance from a node to the spherical surface is as follows: d ═ vt (1-vtcos 45/n °)

Wherein the content of the first and second substances,

v is the model velocity

t is ray propagation time

n is the number of fine divisions.

Since the sphere and the regular polyhedron after division are highly symmetrical graphs, the distances from the finely divided triangular nodes to the sphere are the same.

In step S5, we label the eight triangles of the first divided octahedron 0-7 in the hierarchical labels. In the fine division, each triangle needs to be subdivided into four small triangles. For example, we continue to divide the triangle numbered 1 into numbers (1.0), (1.1), (1.2), and (1.3). Wherein (1.1) is a triangle in the middle, and (1.0), (1.2) and (1.3) are triangles on the left side, the right side and the upper side respectively. If the division is carried out again, the tail mark is added. Therefore, after each fine division, the nodes are corrected to the spherical surface, and only three vertexes of the triangle with the tail mark of 1 need to be corrected.

The invention realizes the wave front construction method by utilizing the triangular gridding rays, and compared with the traditional square grid, the rays are more uniform, have no polar points and are more accurate in grid. Meanwhile, the data are stored and utilized by utilizing the tree-shaped data structure, so that the related calculation efficiency is higher, the known data can be utilized, and rays are added among the rays, so that the calculation is more accurate.

Drawings

FIG. 1 is a schematic diagram of the method of partitioning (first partition, and four fine partitions, respectively);

FIG. 2 is a method of grid numbering;

FIG. 3 is a schematic diagram of a tree structure of a grid;

FIG. 4 is a schematic diagram of filling rays (the distance between the ray 1 and the ray 2 is too large, and the filling distance of the ray 3 is increased);

FIG. 5 is a flow chart of a primary wavefront construction;

FIG. 6 is a flow chart of filling ray intervals;

Detailed Description

The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.

Referring to fig. 5, the present invention provides a method for constructing a wavefront using a triangular gridding ray, which includes the following steps:

and step S2, dividing the grid for the first time and numbering, wherein the grid for the first time is a regular octahedron and has six nodes in total, so that only six rays are provided. As shown in the first graph of fig. 1, where the octahedron is inside the sphere, the error is larger.

In step S3, propagation data calculation is performed on the known six rays.

And step S4, detecting whether the ray density and the distance meet the requirements, and completing primary wave field construction if the ray density and the distance meet the requirements. Otherwise, the mesh is refined once, and for one surface of the octahedron, the octahedron is a regular triangle, and one regular triangle can be divided into four identical small regular triangles, as shown in fig. 2. The division is done for each face of the octahedron to obtain the secondary small triangle data.

In step S5, coordinates of the vertices of the divided small triangle data are corrected. As shown in fig. 2, in the divided small regular triangle (1,1), since all three vertices are inside the sphere, it is necessary to correct the coordinates of the three vertices of the triangle (1,1) so that the three vertices are on the surface of the sphere. Since the sphere and the regular triangle are highly symmetrical graphs, it is proved by geometry that the distances from the vertexes in all the divided secondary triangular spheres to the spherical surface are equal, and the distances are different according to the dividing times, such as the following formula:

fine division for the first time, distance from a node to a spherical surface: d ═ vt (1-cos45 degree)

……

Fine division for the nth time, wherein the distance from a node to the spherical surface is as follows: and d is vt (1-vtcos 45/n DEG), according to a formula, coordinates of the vertex of the triangle are modified, the vertex is on the spherical surface, and therefore, one-time fine division is realized, and the polyhedron is closer to the sphere. We then create a quadtree of the divided triangle nodes for storage, as shown in FIG. 3. Because of the dependency relationship of the tree structure, the whole data structure can be easily searched and traversed. We can use binary tree lookup and traversal. The height of the tree can be obtained through the number of the division, and then index lookup is carried out through the number. In the aspect of traversal, because the height of the tree is known and the number of children under each parent node is known, it is sufficient to traverse the children of each parent node starting from the lowest parent node.

In step S6, the steps S4 and S5 are repeated, increasing the number of rays and increasing the density, so that the tree structure needs to be updated. Since our added data size is computable, i.e. four subdata are added under each of the lowest child nodes, we need only increase the height of the number by 1 for tree structure insertion, and add one more level of data. Wherein the number of data has the following relationship with the number of divisions

The first division (regular octahedron),

the number of nodes is: 6

The number of triangles is: 8

The first time of the fine division is carried out,

the number of nodes is: 6+12 ═ 18

The number of triangles is: 8 x 4-32

The second fine division is carried out on the wafer,

the number of nodes: 6+ 24-30

The number of triangles is: 8 × 4 × 2 ═ 64

……

The n-th fine division is performed,

the number of nodes is: 6+ 12X 2^n-1

The number of triangles is: 8 × 4 × n ═ 64

In the aspect of indexing, only the number of the index needs to be increased, as shown in fig. 2. Meanwhile, in the aspect of coordinate correction, only three fixed points of the middle secondary triangle in each large triangle need to be corrected, namely the triangle with the tail mark of 1.

Since our initialization ray is on a node, and one node of the triangle represents one ray, we also need to traverse the node. Based on the triangle data base, the traversal method comprises the following steps: a computation is performed for all nodes of each triangle, and during the computation we insert a logical variable. Therefore, before each calculation, whether the logic variable exists is checked, if the logic variable does not exist, the calculation is started, and if the logic variable exists, the calculation is skipped.

In step S7, with the ray propagation and the wavefront construction, if the distance between the rays is too large to meet the requirement, one ray is added between the two rays, and propagation data is calculated from the origin, as shown in fig. 4, the wavefront construction is performed four times, and the distance is too large, so the ray 3 is inserted between the ray 1 and the ray 2. In general, the filling ray spacing process is shown in FIG. 6.

Finally, the method of the present application is only a preferred embodiment and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for realizing wavefront construction by utilizing triangular gridding rays is characterized by comprising the following steps:

step S1, obtaining model data, and dividing initial rays by using a triangular mesh;

step S2, in the process of ray propagation, if the distance between the rays is too large, the rays are added between the two rays, and the propagation data of the added rays are calculated from the origin, so as to fill the distance between the rays.

2. The method of claim 1, wherein the triangulating divides the initial ray.

3. The method for filling the ray space of claim 1, wherein a ray is added in the space, and then the propagation data is initialized and calculated till the overlarge space, so as to fill the ray space.