CN115272601A - Method for constructing three-dimensional geological model comprehensive database - Google Patents

Abstract

The application discloses a method for constructing a three-dimensional geological model comprehensive database, which comprises the following steps: acquiring original data of a three-dimensional geological model; extracting boundary information and internal information of the three-dimensional geological model according to the original data to obtain and store a vector grid integrated three-dimensional geometric model of the three-dimensional geological model to form a three-dimensional model data sub-library; acquiring spatial reference system data, standard stratum table data and material data, and respectively establishing a spatial reference coefficient database, a standard stratum table data sub-database and a material data sub-database; and establishing an incidence relation among the three-dimensional model data sub-base, the spatial reference coefficient data sub-base, the standard stratum table data sub-base and the material data sub-base to obtain the three-dimensional geological model comprehensive database. According to the invention, through a three-dimensional geological model vector grid integrated storage mode, a surface-based data object and a volume-based data object can be stored in the same model, the use requirements of different scenes are met, and the three-dimensional geological model can be rapidly read and written.

Description

Method for constructing three-dimensional geological model comprehensive database

Technical Field

The application relates to the technical field of three-dimensional geological modeling, in particular to a three-dimensional geological model comprehensive database construction method.

Background

The three-dimensional geological model is used for restoring spatial distribution characteristics of geological objects as far as possible by applying techniques such as geostatistics, spatial analysis and prediction, geometric reconstruction, computer graphics and the like according to a reliable geological data source. The geological model mainly comprises a face-based three-dimensional geological object model (vector data structure) and a volume-based three-dimensional geological object model (grid data structure) and a mixed data structure mixed model.

In summary, the data model based on the surface representation describes the geometric characteristics of the three-dimensional space entity by using a method of a tiny surface unit or surface element, is good at expressing the outer layer surface structure of the geologic body, can more accurately simulate the surface of an object by encrypting a surface patch, reflects the fluctuation change of the object, has small data volume, can only express the surface structure of the geologic body, and does not pay attention to the internal structure of the geologic body. The three-dimensional geological object model based on the body is mostly based on a voxel model, a complete body model is expressed in a mode of stacking the voxels, a single voxel of a three-dimensional entity can be independently described and stored, and the spatial analysis operation of three-dimensional spatial data is carried out, but the data volume is relatively large, and particularly during visualization, a large amount of useless data possibly exists, and the rendering efficiency of the model is slowed down. The hybrid model integrates the two models, and a vector grid conversion interface is added. The vector data and the grid data are uniformly stored by the hybrid model (vector grid integrated storage model), and a vector grid conversion interface is added, so that the operation difficulty is increased on one hand, and the data precision cannot be ensured on the other hand.

Therefore, the invention provides a three-dimensional geological model comprehensive database construction method, which expresses the internal grid, the external boundary and the geometric topological relation of a geological object by using one model, meets the use requirements of different scenes, can quickly read and write the three-dimensional geological model, and improves the query efficiency of the model.

Disclosure of Invention

In view of the above, it is necessary to provide a method for constructing a three-dimensional geological model integrated database, so as to solve the problems in the prior art that the storage precision of a grid data structure is low, the data size of a vector data structure is large, and the query efficiency is low when modeling source data and modeling results are stored in the field of three-dimensional geological modeling.

In order to solve the above problems, the present invention provides a method for constructing a three-dimensional geological model integrated database, comprising:

acquiring original data of a three-dimensional geological model;

extracting boundary information and internal information of the three-dimensional geological model according to the original data;

obtaining a vector grid integrated three-dimensional geometric model of the three-dimensional geological model according to the boundary information and the internal information, and storing the vector grid integrated three-dimensional geometric model to obtain a three-dimensional model database;

acquiring spatial reference system data, standard stratum table data and material data, and respectively establishing a spatial reference coefficient database, a standard stratum table data sub-database and a material data sub-database;

and establishing the incidence relation among the three-dimensional model data sub-base, the spatial reference coefficient data sub-base, the standard stratum table data sub-base and the material data sub-base to obtain the three-dimensional geological model comprehensive database.

Further, obtaining a vector-grid integrated three-dimensional geometric model of the three-dimensional geological model according to the boundary information and the internal information, including:

obtaining entity boundary data of the three-dimensional geological model according to the boundary information, wherein the entity boundary data is used for storing the outline of the three-dimensional geological model;

obtaining entity network data of the three-dimensional geological model according to the internal information, wherein the entity network data is used for storing geographic characteristic information of the three-dimensional geological model;

and obtaining a vector grid integrated three-dimensional geometric model of the three-dimensional geological model according to the entity boundary data and the entity network data.

Further, the entity boundary data stores the contour of the three-dimensional geological model through four elements of a volume, a surface, a line and a point; each body is surrounded by a group of surfaces, each surface is surrounded by a group of lines, and each line is bounded by two points.

Further, the entity network data stores the geographic characteristic information of the three-dimensional geological model through voxel, patch, edge and vertex; the voxels, patches, edges and vertices form a triangular mesh or a regular grid.

Further, the storage structure of the entity grid data comprises point coordinate data and index information data;

the point coordinate data is used for storing three-dimensional coordinates of all points in the three-dimensional solid model;

the index information comprises vertex index information and structure index information; the vertex index information is used to represent the index of a point, and the structure index information is used to record the vertex index constituting the whole.

Further, the method further comprises:

and obtaining an attribute field model of the three-dimensional geological model according to the internal information data, wherein the attribute field model is used for storing the three-dimensional geological model based on a grid data structure.

Further, the method further comprises:

constructing a geobody element from one or more of the body elements;

forming a geological surface element according to one or more surface elements;

forming a geological line element from one or more of said line elements;

forming geological point elements according to the point elements;

and obtaining a geological three-dimensional solid model according to the geological body elements, the geological surface elements, the geological line elements and the geological point elements, and storing the geological three-dimensional solid model into the three-dimensional model data sub-base.

Further, the method further comprises:

determining attribute data of the vector grid integrated three-dimensional geometric model;

and storing the vector grid integrated three-dimensional geometric model and attribute data corresponding to the model as elements, and performing classified management on the elements according to the attribute data.

Further, the method further comprises:

and storing the elements containing the same characteristic information as an element class.

Further, the method further comprises:

the three-dimensional geological model comprehensive database is established based on a relational database, and data with the reading and writing times larger than a preset reading and writing threshold value is cached by using Redis.

Compared with the prior art, the invention has the beneficial effects that: firstly, extracting boundary information and internal information according to original data of a three-dimensional geological model; secondly, obtaining and storing a vector grid integrated three-dimensional geometric model of the three-dimensional geological model according to the boundary information and the internal information to obtain a three-dimensional model data sub-base; and finally, establishing a spatial reference coefficient database, a standard stratum table database and a material database, and establishing an incidence relation between the spatial reference coefficient database, the standard stratum table database and the material database and the three-dimensional model database to obtain a three-dimensional geological model comprehensive database. According to the invention, through a three-dimensional geological model vector grid integrated storage mode, a surface-based data object and a body-based data object can be stored in the same model, so that the boundary and the internal organization of the three-dimensional geological model can be composed of irregular basic units, the use requirements of different scenes are met, the three-dimensional geological model can be rapidly read and written, and meanwhile, the query efficiency of the model is also improved.

Drawings

FIG. 1 is a schematic flow chart of an embodiment of a method for constructing a three-dimensional geological model comprehensive database according to the present invention;

FIG. 2 is a schematic diagram of an entity boundary data organization according to an embodiment of the body model provided by the present invention;

FIG. 3 is a schematic diagram of an internal mesh data organization of an embodiment of a surface model provided in the present invention;

FIG. 4 is a schematic diagram of a storage architecture of an embodiment of a three-dimensional geosynthetic database provided in the present invention;

FIG. 5 is a diagram illustrating an ordered set key structure according to an embodiment of the present invention.

Detailed Description

The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate preferred embodiments of the invention and together with the description, serve to explain the principles of the invention and not to limit the scope of the invention.

The method is optimized aiming at the problems existing in the storage of the three-dimensional geological model, the current vector data structure can be independently described and stored, and the space analysis operation of the three-dimensional space data is carried out, but the data volume is relatively large, and particularly in visualization, a large amount of useless data possibly exists, and the rendering efficiency of the model is slowed down; the grid data is good at expressing the outer layer surface structure of the geologic body, the surface of the object can be simulated more accurately by encrypting the patch, the fluctuation change of the object is reflected, the data volume is small, but only the surface structure of the geologic body can be expressed, and the internal structure of the geologic body is not concerned. The invention provides a boundary and voxel coupled body model organization method, which stores a surface-based data object and a body-based data object in the same model, so that the boundary and the internal organization of a three-dimensional geological model can be composed of irregular basic units, different use requirements are met, the three-dimensional geological model can be rapidly read and written, and the query efficiency of the model is improved.

The embodiment of the invention provides a method for constructing a three-dimensional geological model comprehensive database, and fig. 1 is a flow diagram of the method for constructing the three-dimensional geological model comprehensive database, which comprises the following steps:

step S101: acquiring original data of a three-dimensional geological model;

step S102: extracting boundary information and internal information of the three-dimensional geological model according to the original data;

step S103: obtaining a vector grid integrated three-dimensional geometric model of the three-dimensional geological model according to the boundary information and the internal information, and storing the vector grid integrated three-dimensional geometric model to obtain a three-dimensional model database;

step S104: acquiring spatial reference system data, standard stratum table data and material data, and respectively establishing a spatial reference coefficient database, a standard stratum table data sub-database and a material data sub-database;

step S105: and establishing the incidence relation among the three-dimensional model data sub-base, the spatial reference coefficient data sub-base, the standard stratum table data sub-base and the material data sub-base to obtain the three-dimensional geological model comprehensive database.

According to the method for constructing the three-dimensional geological model comprehensive database, firstly, boundary information and internal information are extracted according to original data of a three-dimensional geological model; secondly, obtaining and storing a vector grid integrated three-dimensional geometric model of the three-dimensional geological model according to the boundary information and the internal information to obtain a three-dimensional model data sub-base; and finally, establishing a spatial reference coefficient database, a standard stratum table database and a material database, and establishing an incidence relation between the spatial reference coefficient database, the standard stratum table database and the material database and the three-dimensional model database to obtain a three-dimensional geological model comprehensive database. In the embodiment, the data object based on the surface and the data object based on the body can be stored in the same model in a way of integrally storing the vector grid of the three-dimensional geological model, so that the boundary and the internal organization of the three-dimensional geological model can be composed of irregular basic units, the use requirements of different scenes are met, the three-dimensional geological model can be rapidly read and written, and meanwhile, the query efficiency of the model is also improved.

As a specific example, the spatial reference coefficient database comprises a plurality of different coordinate systems for determining coordinate system information of the three-dimensional geological model.

As a specific embodiment, the standard stratum table data sub-database records stratum sequence, and can simply and intuitively display the upper and lower stratum relations of the three-dimensional geological model, thereby facilitating stratum comparison and explaining the chronological relation of the stratum sequence.

As a specific embodiment, the material database is used for recording material information of the three-dimensional geological model, and is linked to the specific three-dimensional geological model for association by recording information such as material name and material file path. The material database includes color database, texture database, color table database, etc.

As a preferred embodiment, in step S103, obtaining a vector-grid integrated three-dimensional geometric model of the three-dimensional geological model according to the boundary information and the internal information, including:

obtaining entity boundary data of the three-dimensional geological model according to the boundary information, wherein the entity boundary data is used for storing the outline of the three-dimensional geological model;

obtaining entity network data of the three-dimensional geological model according to the internal information, wherein the entity network data is used for storing geographic characteristic information of the three-dimensional geological model;

and obtaining a vector grid integrated three-dimensional geometric model of the three-dimensional geological model according to the entity boundary data and the entity network data.

Storing the external boundary data separately may allow faster retrieval of the boundary data for subsequent calculations, visualizations, and the like.

In a preferred embodiment, the entity boundary data stores the contour of the three-dimensional geological model through four elements of a volume, a surface, a line and a point; each body is formed by a group of surfaces in a surrounding mode, each surface is formed by a group of lines in a surrounding mode, and each line is bounded by two points.

The above-described storage method is described below with reference to fig. 2 by using a specific embodiment.

FIG. 2 is a body model consisting of top and bottom surfaces plus three side surfaces, each surface consisting of three or four lines, each line having two nodes, the underlying structure forming the boundary of the overlying structure.

When an individual is completely represented, nodes, lines, surfaces and the individual need to be stored in sequence, but only the outer boundary of the three-dimensional geometric model is expressed in the mode, and in order to express the three-dimensional geometric model more accurately, the internal grid data also needs to be stored. Taking a simple side as an example, the outer boundary of the side can be regarded as a polygon, the polygon can only express the outline of the side, but cannot show the internal condition thereof, and internal mesh data, such as a triangular mesh, a regular mesh, etc., needs to be stored.

As a preferred embodiment, the entity network data stores the geographic feature information of the three-dimensional geological model through four elements of a voxel, a patch, an edge and a vertex; the voxels, patches, edges and vertices form a triangular mesh or a regular grid.

As a preferred embodiment, the storage structure of the entity mesh data includes point coordinate data and index information data;

the point coordinate data is used for storing three-dimensional coordinates of all points in the three-dimensional solid model;

the index information comprises vertex index information and structure index information; the vertex index information is used to indicate an index of a point, and the structure index information is used to record vertex index information constituting the entire structure.

The above-described physical grid data storage structure is explained below with reference to fig. 3.

The surface model shown in fig. 3 is composed of two surface patches a and B, and there are five vertices in total, where vertex 1 is the origin, the coordinates are (0, 0), and the Z coordinates of all vertices are uniformly marked as 0.

The vertex data of the surface model sequentially records three-dimensional coordinates of five vertexes, the vertex indexes of the surface patches record vertex index numbers of a surface A and a surface B, the surface A consists of four points of 0,1,2 and 3, and the surface B consists of three points of 3, 2 and 4. But the patch vertex index does not record which points make up a face, and this information is recorded by the structure index.

The structure index data records index information, but the structure index is an index in the vertex index table of the record patch, and the start and stop indexes of all patches are recorded in the item information. The start-stop index interval of the A surface is 0-3, the start-stop index interval of the B surface is 4-6, and if the third surface exists, the start index of the B surface is 7.

From the above analysis, it can be found that the data number of the point coordinate data table is three times of the vertex number, in the index information table, the data number of the vertex index table is the sum of the number of all the surface patches, and the data number of the surface patch data table is the sum of the surface patch number and one.

By separating the index information of the patch into the vertex index and the structure index, not only models of different types of patches can be compatible, but also different types of patches can be used in the same model.

Similarly, for a volume model, a voxel is the basic unit of a volume. The internal mesh data structure of the volume model is similar to the surface model. The coordinate information of all the vertexes is recorded in a point coordinate data table, the vertex index of each voxel is recorded in the vertex index of each voxel, and the structural data of the voxel is recorded in which vertexes constitute the voxel. The internal mesh data structure of the volume model may be compatible with models of different types of voxels, or different types of voxels may be used in the same volume model.

In addition, to reduce the complexity of the associative table lookup, each three-dimensional geometric model independently stores its own internal mesh data structure. The internal mesh data already contains the external boundary data, and it is necessary to store part of the data repeatedly. It should be noted that the external boundary data is not set as a necessary option, and whether the data needs to be stored can be determined according to actual needs.

As a specific embodiment, the point database table, the line database table, the surface database table, and the volume database table are designed to store the point geological model, the line geological model, the surface geological model, and the volume geological model respectively.

(1) Point database table

The Point database table mainly records the geometric information of the local points, and for the convenience of representation, the vertex coordinates are modified by Point type, namely three-dimensional space coordinates (x, y, z).

Table 1 Point database table

Numbering	Name of field	Field type	Description of field
				1	cornerID	INTEGER	Node ID
2	cornerDATE	DATE	Time stamp
				3	cornerExist	INTEGER	Presence sign
4	matType	INTEGER	Type of material
				5	matName	TEXT	Name of material
6	vertData	POINT(3)	Vertex coordinates

(2) Thread

The line database table records basic information such as geometry and line parameters of a line in units of one line.

Table 2 line database table

Numbering	Name of field	Type of field	Description of field
				1	LinID	INTEGER	Line ID
2	linDATE	DATE	Time stamp
				3	linExist	INTEGER	Presence sign
4	LinBOX	BOX	Geometric range
				5	linBound	BLOB	Line boundaries
6	vertData	BLOB	Vertex data
				7	linEdgeData	BLOB	Side data
8	LineWidth	REAL	Line width
				9	matType	INTEGER	Type of material
10	matName	TEXT	Name of material

Description of related fields:

1. line boundary: the field records only the ID information of the head and tail nodes of the node list of the end points of the line, two nodes per line.

Data organization of line boundaries: node 1ID, node 2ID.

2. Vertex data: number of vertices, coordinates of all points on the line.

Data organization of vertex data: the number n of vertices, and color, texture, and coordinate information of the n vertices are sequentially arranged.

3. Side data: number of edges, list of vertex numbers of all edges.

Data organization form of the side data: the number of the edges n, and indexes of start and stop vertexes of the n edges are sequentially arranged.

(3) Surface database table

The surface is the most basic unit for storing a model, the geometric information of the model is mainly stored in the table, and the surface has internal and external attributes, so the material information is represented by internal and external fields. By separately recording the vertex data, patch vertex data, and patch data, storage can be performed without limitation on patch types (different polygons can be allowed to coexist).

Table 3 surface database table

Numbering	Name of field	Field type	Description of field
				1	surfID	INTEGER	Noodle ID
2	surfDATE	DATE	Time stamp
				3	surfExist	INTEGER	Presence sign
4	surfBOX	BOX	Geometric range
				5	SurfBound	BLOB	Boundary of surface
6	VertData	BLOB	Vertex data
				7	Facetvertices	BLOB	Patch vertex data
8	Facetptr	BLOB	Patch data
				9	adjFacetData	BLOB	Contiguous patch data
10	InMatType	INTEGER	Type of material
				11	InMatName	TEXT	Name of material
12	OutMatType	INTEGER	Type of material
				13	OutMatName	TEXT	Name of material

Description of related fields:

1. surface boundary: the number of lines constituting a surface, and a line ID list.

Data organization of the face boundaries: the number of lines n, + line 1ID, -line 2ID, \ 8230, + line nID (+ indicating clockwise, -number indicating counterclockwise).

2. Vertex data: i.e. the geometrical information of points in the surface network.

Data organization of vertex data: the number n of vertices, and color, texture, and coordinate information of the n vertices are sequentially arranged.

3. Patch vertex data: the number of patches and a list of vertex numbers of the patches.

Data organization form of patch vertex data: the number of patches m,0,1,2, \ 8230, n-1 (here, the vertex index, the maximum value does not exceed n-1).

4. Patch data: index information of each patch is recorded.

The data organization form of the patch data is 0,3,6 \8230 (here, it is a triangle, i.e. 0,1,2 is a sub-surface, 3, 4, 5 is a sub-surface).

5. Contiguous patch data: number of adjacent patches, list of vertex numbers of adjacent patches.

Data organization of contiguous patch data: the number n of contiguous patches, the start and stop index numbers of the n patches.

(4) Volume database table

The body surface is formed into a boundary by multiple surfaces, and the inside of the body surface can be filled with voxels, and the topological information of the body surface is recorded.

Table 4 volume database table

Number of	Name of field	Type of field	Description of field
				1	RegionID	INTEGER	Body ID
2	RegionDATE	DATE	Time stamp
				3	RegionExist	INTEGER	Presence sign
4	RegionBOX	BOX	Geometric range
				5	RegionBound	BLOB	Body boundary
6	vertData	BLOB	Vertex data
				7	cellVertData	BLOB	Voxel vertex index list
8	CellFacetData	BLOB	Voxel data
				9	adjCellVertData	BLOB	Contiguous voxel data
10	matType	INTEGER	Type of material
				11	matName	TEXT	Name of material

Description of related fields:

1. body boundary: the number of face boundaries constituting the construct, and the ID list of the faces.

The number of faces n, + face 1ID, -face 2ID, \ 8230, + face nID (+ denotes front face, -number denotes back face).

2. Vertex data: and (4) listing the vertex number and the vertex coordinates of the voxel.

Data organization of vertex data: the number n of vertices, and the color, texture, and coordinate information of the n vertices are sequentially arranged.

3. Voxel vertex index data: voxel number and voxel vertex index list.

Data organization of voxel vertex index data: the number of voxels m,0,1,2, \ 8230n-1 (here the vertex index, the maximum value does not exceed n-1).

4. Voxel data: indices of the first vertex and the last vertex of each voxel.

The data organization of the voxel data (BLOB) is 0,4,8 \8230 (here tetrahedron, i.e. 0,1,2, 3 constitute one voxel, 4, 5, 6, 7 constitute one voxel).

5. Contiguous voxel data: the number of adjacent voxels and an adjacent voxel number list.

Data organization of contiguous voxel data:

the number n of contiguous voxels, the start-stop index number of the n contiguous voxels.

As a preferred embodiment, the method further comprises:

constructing a geobody element from one or more of the body elements;

forming a geological surface element according to one or more surface elements;

forming a geological line element from one or more of said line elements;

forming geological point elements according to the point elements;

and obtaining a geological three-dimensional solid model according to the geological body elements, the geological surface elements, the geological line elements and the geological point elements, and storing the three-dimensional solid model into the three-dimensional model data sub-base.

The geological three-dimensional solid model is an extension on the four types of geometric elements, such as a geological point which is composed of one or more geometric points and adds related geological significance.

As a specific embodiment, the geological point elements include two types, wherein the first type is a common geological point and description information of the common geological point is recorded; the second type is the occurrence, and besides the description information, the tendency and inclination angle of the occurrence are also recorded. Geologic lines, geologic surfaces, and geologic bodies, all of which are similar.

The number of geometric elements in the geological three-dimensional entity is not fixed, and due to the complexity of the geological entity, the adjacent relation between the geometric entities does not need to be recorded.

As a preferred embodiment, the method further comprises:

and obtaining an attribute field model of the three-dimensional geological model according to the internal information data, wherein the attribute field model is used for storing the three-dimensional geological model based on a grid data structure.

As a specific embodiment, the attribute field model includes attributes corresponding to a geometry storage model of the three-dimensional geological model based on a raster data structure, and LOD data, where the LOD data is used to display the geometry storage model with different accuracies according to requirements.

As a preferred embodiment, the method further comprises:

determining attribute data of the vector grid integrated three-dimensional geometric model;

and storing the vector grid integrated three-dimensional geometric model, the attribute data corresponding to the geometric model and the attribute field model of the three-dimensional geological model as elements, and carrying out classified management on the elements according to the attribute data.

As a preferred embodiment, the method further comprises:

and storing the elements containing the same characteristic information as an element class.

As a specific example, the technical solution is shown in conjunction with fig. 4. As shown in fig. 4, the three-dimensional geological model integrated database includes a three-dimensional model data sub-base, a spatial reference coefficient data sub-base, a standard formation table data sub-base and a material data sub-base. The three-dimensional model data sub-library comprises a plurality of element classes, and the three-dimensional model data sub-library is classified and managed through the element classes. The element class includes a plurality of elements containing the same characteristic information. The elements comprise a vector grid integrated three-dimensional geometric model, attribute data corresponding to the model and an attribute field model of the geological three-dimensional model.

Therefore, the basic frame structure of the whole storage engine is designed according to the composition mode of a three-dimensional geological model comprehensive database-a three-dimensional model data sub-database-an element class-an element-an entity (a geometric model + attribute data + attribute field model).

As a specific embodiment, the three-dimensional model data sub-library further includes a three-dimensional legend and a three-dimensional label, and the three-dimensional legend and the three-dimensional label are respectively associated with the element classes.

As a specific embodiment, the three-dimensional annotation comprises two types, namely a static annotation and an attribute annotation. The static annotation has a fixed three-dimensional coordinate position, the attribute annotation content is from the attribute values of a plurality of fields of the element, and the annotation is associated with the element to be annotated when displayed.

As a specific example, the storage process for any one three-dimensional geological model is as follows:

1. analyzing the three-dimensional geological model and determining the element class of the model;

2. determining the elements contained in the element classes and the total amount of the elements contained in each element class;

3. writing the element data into a corresponding element table;

4. and storing the data of the vector grid integrated three-dimensional geometric model of the three-dimensional geological model.

In order to quickly read and write the model, the organization form of multi-source heterogeneous geological data and a three-dimensional model and a reasonably designed database table storage structure need to be combed, so that the storage space and the calculation power can not be wasted on the data organization. As will be described in detail below.

First, in terms of caching of the database:

as a preferred embodiment, the three-dimensional geological model comprehensive database is established based on a relational database, and data with the reading and writing times larger than a preset reading and writing threshold value is cached.

Specifically, in order to store data persistently, the embodiment uses a relational database to establish a three-dimensional geological model comprehensive database; and caching the hot spot data by adopting Redis. In addition, the present embodiment optimizes the cache granularity, the storage structure, the cache update policy, and the cache eviction policy.

(1) Cache granularity design

There are many factors that affect the actual occupied capacity of the cache, and the granularity of the cache is one of them. The cache granularity refers to the minimum unit of cache information, or the hierarchy of cache information blocks. When the cache granularity is large, cache hits are more difficult. Conversely, when the cache granularity is smaller, cache hits are more likely to occur, but this also means more effort and complexity.

For a three-dimensional geological model, if the granularity is defined as the three-dimensional model, one record in the cache will be large and will contain the whole content of the three-dimensional model, and a large amount of redundancy is generated in practical application, which is not preferable.

The embodiment adopts a mode of packaging frequently used SQL statements as functions and taking a database interface as cache granularity. When the same database interface function is used again and the parameters are the same, a cache hit is triggered. In this way, the SQL statements to be processed need to be classified in advance, and encapsulated as different functions, and information such as table names and field names is recorded by function parameters.

According to the application scene analysis of the three-dimensional geological model, the hot spot data is concentrated in the structural data of the three-dimensional model, and the storage design is storage facing to the geological object, so that the hot spot data objects can be cached, such as a surface object, a body object and the like.

(2) Cache memory structure design

Redis is a key-value type in-memory database, and a key is used as a unique identifier of a record and is firstly ensured not to be repeated. And storing different geological objects in different tables in a three-dimensional geological model storage library, and taking the ID number as a unique identifier. The key can be designed in such a way that the object name + ID number is used. For example, when one surface is cached, the key may be surface _1, which indicates the surface with ID number 1. For less memory space, the object name can also be defined as a numerical value in advance, for example, 1 represents a surface, and 2 represents a body, so that the key value of the surface model in fig. 3 is 1 \u1.

Redis supports data structures with various characteristics such as character strings, hashes, lists, sets and the like, and is a basic form for storing actual data, so that a proper structure needs to be selected according to the characteristics of three-dimensional geological model data. The present embodiment uses a String structure, i.e., string type, for the following reasons: firstly, most of model hotspot data are triangular net or regular data, serialization and deserialization are easy to perform, and serialization is inevitably performed when BLOB fields are used for storage; and the String type can be regarded as a record which only has one field except the key value, so that the space is saved, and the query is convenient. The biggest drawback of this is that when a certain data change is needed for a record, the whole record needs to be modified, which has a great influence on the efficiency. For a three-dimensional geological model, data is generally not modified for a certain area of a certain surface, and the modification is mostly for the whole model, so the influence of the defects can be ignored.

(3) Cache update policy design

The essence of the cache is to place a same block of data in the high-speed device, so that the application only needs to access the cache in the high-speed device and does not need to access the actual data of the low-speed device when accessing the data block, which requires that the data in the high-speed device and the data in the low-speed device are consistent, and the wrong data cannot be extracted due to the efficiency improvement, which requires that a proper cache updating strategy is designed to ensure the data consistency.

In the field of geological modeling, the processes of modeling data preprocessing, modeling algorithm model construction, model visualization, warehousing and the like are followed, visualization is performed after the model construction is successful, and modeling parameters or data are adjusted and modeling is performed again when the model construction is not successful. Only if the model is in good agreement with the expectation is the model put in storage. Therefore, most operations on the geological model database are concentrated on query, and all actual data of geological objects are serialized into a String type field in the storage structure design of the cache, so that the scheme uses the following cache updating mode, when the data needs to be updated, the data is updated into the database, and the cache data is deleted after the data is successfully updated. The cache hit or miss logic in this mode is as follows: the application program firstly acquires data from the cache layer and directly returns the data after the data is successfully acquired; and if the data is failed, acquiring the data from the database, and after the data is successfully acquired, writing the data into the cache and returning.

(4) Cache eviction policy design

Because the memory capacity is limited, the cache based on the memory database cannot place cache data infinitely, and a maximum capacity value is inevitably set in order to ensure the normalized operation of the whole environment. When the cache data reaches the capacity value, some cache needs to be deleted by using a certain strategy to ensure that the cache system can continue to operate, which is the meaning of designing cache eviction.

In addition, the efficiency can be better ensured by setting the expiration time for the data. Even if the cache has not reached the set maximum usage capacity, the data may be automatically deleted because the time has expired.

The commonly used cache eviction algorithms are mainly FIFO, LRU, LFU, etc. For geological model storage, hot data is often concentrated in a certain model, and generally, the distribution of the hot data has no direct relation with time, so that the LFU algorithm is the algorithm most suitable for a three-dimensional geological model storage cache eviction strategy. The present embodiment uses the LFU algorithm, but integrates other algorithms, and an external interface for modifying the eviction algorithm may be designed to adapt to different application scenarios.

In the aspect of indexing technology:

the given spatial range query data is a common spatial query in the field of geological modeling. For example, in three-dimensional geological modeling based on drilling data, point positions of a drilling database are firstly spread, a modeling space range is given by inputting data or in a man-machine interaction frame drawing mode, and modeling is finally carried out based on the drilling data and other source data in the space range.

The spatial index technology is used for screening the drilling data in the modeling spatial range, and the spatial index solution is provided for the cache layer and the database layer respectively.

In a cache layer, based on a Redis memory database, two spatial index query schemes are designed by summarizing and summarizing application scenes of geological model database spatial query:

firstly, model data (latest query) near a certain point is queried and solved through a GeoSpatial module provided by Redis;

secondly, constrained polygon query (range query), if the range query takes out the selected model data, the model data is solved through the sorted set data type, and the following detailed description is given:

(1) Recent queries

Redis provides a data type-GeoSpatial capable of recording geographic positions, the names and the geographic positions of data items can be added into specified keys, the GeoHash arrangement is completed inside the data items, and then the data items can be used for computing distance, nearby element screening and the like.

For geological entities needing to be added with spatial indexes, different key keys can be created according to names of the geological entities, the key keys are not influenced mutually, and the names of the data items directly use unique identification ID numbers. Therefore, after the warehousing operation is completed, a spatial index is constructed in the cache according to the ID number and the position of the spatial index.

The GeoHash algorithm used by the data type actually adopts the idea of dimension reduction, two-dimensional or high latitude data coordinates are converted into one-dimensional character strings, and the nearest elements can be obtained by comparing whether the character strings are similar or not. Therefore, the query efficiency is high, and the method is widely applied to map industry. By using the data type in a cache layer of the three-dimensional geological model storage, the efficiency of recent queries can be improved.

(2) Range query

Redis does not support the direct index establishment based on the element BOX, but can establish a sequencing structure for the coordinates of the element BOX by using a Z-SET data index, and can meet the requirement of range query under the condition of improving the efficiency.

As a key-value type memory database, redis also supports types such as SET (collection), ordered SET (ordered collection), etc., and when a field type is a numerical value, an ordered collection can be used to build an ordering index. Similar to a collection, the strings in an ordered collection are also unique and non-repeatable, but each member of the ordered collection is bound to a score (score) and ordered by the size of the value. Therefore, data items with scores in a certain interval can be accessed quickly, and the data items do not need to be accessed and screened in sequence.

Taking drilling data as an example, the hole site coordinates are three-dimensional coordinates, when practical engineering application interaction is carried out, the drilling data can be projected into a two-dimensional view, drilling holes needing modeling are selected through frame pulling interaction, and then subsequent operations such as modeling, visualization, warehousing and the like are carried out. At this time, only latitude and longitude coordinates of the borehole are noted, and represented by XY. Thus, four ordered sets minX, maxX, minY, maxY can be established. As shown in fig. 5, the score of each sorted set is a coordinate value of the stored element, and the actual ID of the element is recorded.

And after the drilling data are stored, creating the ordered set in the cache according to the point location information and the ID value of the drilling data. When the range query is needed, ZRANGEBYSCORE operation is respectively carried out on the four ordered sets to obtain four query results, and finally an intersection is taken to obtain a final result, so that the efficient query based on the BOX index is completed.

At the database level, the R-tree index is used. When the data volume is not very large and the concurrency is not large, only the database index can be opened; when the amount of data and the amount of concurrency are large, a cache level spatial index may be used. The database spatial index relies on a virtual table mechanism, and the cache layer spatial index does not need any processing for the database. Therefore, the switching of the two types of spatial indexes can be regarded as the handling of data in two tables in the database so as to ensure the uniqueness and integrity of the data.

When the tables are built, the two tables are built at the same time, but only one type of the tables is used at the same time. When the space index of the database is switched into the space index of the cache layer, reading data from the virtual table, inserting the data into a corresponding common database table, emptying the data of the virtual table, and finally reading the ID and the coordinate information of the virtual table to construct the space index in the cache layer; when the cache layer spatial index is switched to the database spatial index, reading data from the common database table, inserting the data into the corresponding virtual tables one by one, then emptying the common database table, and finally deleting the cache layer spatial index.

The specific query steps are as follows:

(1) and (4) click selection: clicking the screen, calculating the coordinate position and the buffer area range by the system, preferentially matching the data objects in the cache layer, if the matching is not successful, performing line matching in the database, and finally returning object information.

(2) Selecting frames: and (4) framing the range, calculating the query range by the system, wherein the query mechanism is consistent with the click query, and only the range result is that all element object information in the framing area is returned.

(3) And (3) keyword retrieval: inputting keywords, preferentially matching the data information of the cache layer, and returning all object information meeting the requirements if the matching is successful; and if the matching is unsuccessful, matching corresponding field information of the database, and returning all successfully matched object information.

The invention discloses a method for constructing a three-dimensional geological model comprehensive database, which comprises the following steps of firstly, extracting boundary information and internal information according to original data of a three-dimensional geological model; secondly, obtaining and storing a vector grid integrated three-dimensional geometric model of the three-dimensional geological model according to the boundary information and the internal information to obtain a three-dimensional model database; and finally, establishing a spatial reference coefficient data sub-base, a standard stratum table data sub-base and a material data sub-base, and establishing an incidence relation between the spatial reference coefficient data sub-base, the standard stratum table data sub-base and the material data sub-base and the three-dimensional model data sub-base to obtain a three-dimensional geological model comprehensive database.

According to the invention, through a three-dimensional geological model vector grid integrated storage mode, a surface-based data object and a volume-based data object can be stored in the same model, so that the boundary and the internal organization of the three-dimensional geological model can be composed of irregular basic units, different use requirements are met, the three-dimensional geological model can be rapidly read and written, and meanwhile, the query efficiency of the model is also improved.

While the invention has been described with reference to specific preferred embodiments, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the following claims.

Claims (10)

1. A method for constructing a three-dimensional geological model comprehensive database is characterized by comprising the following steps:

acquiring original data of a three-dimensional geological model;

extracting boundary information and internal information of the three-dimensional geological model according to the original data;

obtaining a vector grid integrated three-dimensional geometric model of the three-dimensional geological model according to the boundary information and the internal information, and storing the vector grid integrated three-dimensional geometric model to obtain a three-dimensional model database;

acquiring spatial reference system data, standard stratum table data and material data, and respectively establishing a spatial reference coefficient database, a standard stratum table data sub-database and a material data sub-database;

and establishing the incidence relation among the three-dimensional model data sub-base, the spatial reference coefficient data sub-base, the standard stratum table data sub-base and the material data sub-base to obtain the three-dimensional geological model comprehensive database.

2. The method for constructing the three-dimensional geological model comprehensive database according to claim 1, wherein obtaining a vector-grid integrated three-dimensional geometric model of the three-dimensional geological model according to the boundary information and the internal information comprises:

obtaining entity boundary data of the three-dimensional geological model according to the boundary information, wherein the entity boundary data is used for storing the outline of the three-dimensional geological model;

obtaining entity network data of the three-dimensional geological model according to the internal information, wherein the entity network data is used for storing geographic characteristic information of the three-dimensional geological model;

and obtaining a vector grid integrated three-dimensional geometric model of the three-dimensional geological model according to the entity boundary data and the entity network data.

3. The method for constructing the three-dimensional geological model comprehensive database according to claim 2, wherein the entity boundary data stores the contour of the three-dimensional geological model by four elements of a body, a face, a line and a point; each body is surrounded by a group of surfaces, each surface is surrounded by a group of lines, and each line is bounded by two points.

4. The method for constructing a three-dimensional geological model comprehensive database according to claim 2, wherein the entity network data stores the geographic characteristic information of the three-dimensional geological model by four elements of voxel, patch, edge and vertex; the voxels, patches, edges and vertices form a triangular mesh or a regular grid.

5. The method of constructing a three-dimensional geological model integrated database according to claim 4, wherein the storage structure of the solid grid data comprises point coordinate data and index information data;

the point coordinate data is used for storing three-dimensional coordinates of all points in the three-dimensional solid model;

the index information comprises vertex index information and structure index information; the vertex index information is used to represent the index of a point, and the structure index information is used to record the vertex index constituting the whole.

6. The method of building a three-dimensional geological model integrated database according to claim 1, further comprising:

and obtaining an attribute field model of the three-dimensional geological model according to the internal information data, wherein the attribute field model is used for storing the three-dimensional geological model based on a grid data structure.

7. The method of constructing a three-dimensional geological model integrated database according to claim 6, further comprising:

determining attribute data of the vector grid integrated three-dimensional geometric model;

and storing the vector grid integrated three-dimensional geometric model, the attribute data corresponding to the geometric model and the attribute field model of the three-dimensional geological model as elements, and carrying out classified management on the elements according to the attribute data.

8. The method of building a three-dimensional geological model integrated database according to claim 7, further comprising:

and storing the elements containing the same characteristic information as an element class.

9. The method of building a three-dimensional geological model integrated database according to claim 1, further comprising:

constructing a geobody element from one or more of the body elements;

forming geological surface elements according to one or more surface elements;

forming a geological line element from one or more of said line elements;

forming geological point elements according to the point elements;

and obtaining a geological three-dimensional solid model according to the geological body elements, the geological surface elements, the geological line elements and the geological point elements, and storing the geological three-dimensional solid model into the three-dimensional model data sub-base.

10. The method for constructing the three-dimensional geological model comprehensive database according to claim 1, further comprising:

the three-dimensional geological model comprehensive database is established based on a relational database, and data with the reading and writing times larger than a preset reading and writing threshold value is cached by using Redis.

Images