CN116863097A

CN116863097A - Space environment data model construction method and device

Info

Publication number: CN116863097A
Application number: CN202311134675.0A
Authority: CN
Inventors: 杨婧; 王宇翔; 王涛; 柏冰
Original assignee: Aerospace Hongtu Information Technology Co Ltd
Current assignee: Aerospace Hongtu Information Technology Co Ltd
Priority date: 2023-09-05
Filing date: 2023-09-05
Publication date: 2023-10-10

Abstract

The application provides a method and a device for constructing a space environment data model, and relates to the technical field of data model construction, wherein the method comprises the following steps: extracting space multi-dimensional element information based on an application scene, and performing digital twin classification processing on the space multi-dimensional element information to obtain space environment multi-modal data; modeling a spatial data model according to the spatial environment multi-mode data to obtain a spatial environment data model; determining the association relation among the data models based on element information, a description method, data types and data organization among the data models in the spatial environment data model; mapping the spatial environment data model to a spatial environment entity; the space environment entity is in a triplet form formed by an entity structure, attribute elements and association relations. The application constructs a space environment data model covering the physical object, phenomenon, event and other information of the whole space and all elements of the whole domain, and designs the mapping rule from the multi-mode data model to the space environment entity.

Description

Space environment data model construction method and device

Technical Field

The application relates to the technical field of data model construction, in particular to a method and a device for constructing a space environment data model.

Background

The construction of the space environment data model can lay a theoretical foundation for the construction of space environment entities and support the application of twin space, such as urban planning, disaster monitoring, logistics management and the like. In the related art, a GIS data model is usually adopted, however, the GIS data model cannot meet the requirement of incomplete characterization of the GIS data model along with the expansion of the application of a spatial information system, and thus inaccurate model results are caused. In addition, the GIS data model modeling process is relatively fractured, the requirement of describing the characteristics of the same entity object by using different time-space scale data is difficult to meet, and the association between data models is not considered.

Disclosure of Invention

The application aims to provide a space environment data model construction method and device, a space environment data model which covers the information of physical objects, phenomena, events and the like of all-space and all-elements of a whole domain is constructed, and a mapping rule from a multi-mode data model to a space environment entity is designed.

In a first aspect, the present application provides a method for constructing a spatial environment data model, including: extracting space environment multi-dimensional element information based on an application scene, and carrying out digital twin classification on the space environment multi-dimensional element information to obtain multi-mode space environment data; modeling a spatial data model according to the spatial environment multi-mode data to obtain a spatial environment data model; the space environment data model comprises a grid model, a scene model, a voxel vector model, a field model, a flow model, a special effect model, a physical model, a chemical model, a complex efficiency model, a facility model, a relation function model and an event map model; determining the association relation among the data models based on element information, a description method, data types and data organization among the data models in the spatial environment data model; mapping the spatial environment data model to a spatial environment entity; the space environment entity is in a triplet form formed by an entity structure, attribute elements and association relations.

In an alternative embodiment, modeling the spatial data model according to the spatial environment multi-mode data to obtain a spatial environment data model, including: identifying each grade of subdivision grids by adopting 64-bit coding aiming at the grid model; selecting a corresponding grid value method based on the type of the grid element attribute to carry out modeling, and constructing a grid model; the grid element attribute at least comprises population, economy, weather and hydrology, and the value of each grid unit of the grid model is determined by the area attribute value corresponding to the grid center point.

In an alternative embodiment, modeling the spatial data model according to the spatial environment multi-mode data to obtain a spatial environment data model, including: obtaining real projective image data, labeling the real projective image data with a geographic entity, establishing a training sample set, and performing image interpretation on the training sample set to obtain the geographic entity expressed in a two-dimensional form; selecting an oblique photography three-dimensional model based on the acquisition precision of the ground object entity, then cutting a triangular net and texture to obtain a three-dimensional geometric shape and texture, and obtaining a three-dimensional form expressed geographical entity through preset restoration processing; and constructing a semantic description framework, acquiring semantic information, constructing a hierarchical relationship, and generating a voxel vector model.

In an alternative embodiment, modeling the spatial data model according to the spatial environment multi-mode data to obtain a spatial environment data model, including: establishing a facility attribute database, and calling facility component geometric structure data based on the voxel vector model; establishing a corresponding relation between a facility attribute database and a voxel vector model structure to carry out semantic attribute association, acquiring characteristic points of facilities according to the voxel vector model structure, calculating attribute values of the characteristic points, and constructing a facility model.

In an alternative embodiment, the field model includes a gravitational field model, a geomagnetic field model, a cloud field model, a wind field model, and a fog field model; modeling the spatial data model according to the spatial environment multi-mode data to obtain a spatial environment data model, including: calculating a gravity quasi-geoid by a removal-recovery technology, and constructing a gravity field model; calculating a global model of the geomagnetic field by adopting a spherical harmonic analysis method, and calculating a regional model of the geomagnetic field by adopting a crown harmonic analysis method; a cloud field model is built by calculating three-dimensional cloud cover, characteristic layer cloud cover, cloud bottom height and cloud top height; decomposing wind field data into an intrinsic mode and a main coordinate through a characteristic value by adopting an intrinsic orthogonal decomposition method, and constructing a wind field model; and constructing a fog field model by solving the average visibility value in each grid.

In an alternative embodiment, for the physical model, a resource library is first established according to a physical formula, an influence coefficient, a symbol, and the like; correlating a physical formula with the related attribute of the voxel vector model to acquire physical state change information when the entity contacts; the physical formula is converted into a general formula of a physical model by using a numerical simulation method, and the physical model has the following unified expression:

wherein i represents a physical quantity contained in the model, the physical quantity at least comprises speed, time, distance and quality in the mechanical model,for influencing the coefficient, the influencing coefficient at least comprises a dynamic influencing coefficient, a friction coefficient, a resistance coefficient and a heat conduction coefficient, and y represents the value of the physical model.

In an optional embodiment, for the event map model, based on network resource data, performing event identification by adopting a method of jointly identifying an entity and an event trigger word based on sequence labeling, extracting event elements in an event description sentence according to a predefined event representation frame, and judging event types of the event elements, wherein the event types at least comprise event occurrence types, event influence types and emergency decision types; constructing candidate event pairs, carrying out standardization processing on structured event elements, describing the similarity of event types through the similarity of the three aspects of entities, the event elements and the event semantics, and solving the problem of event element conflict possibly generated in the completion process; judging event correlation by adopting a co-occurrence event method, counting occurrence probability among events with correlation, mapping the marked relationship in an event corpus to event relationship, and establishing an event knowledge graph; and visualizing the event knowledge graph by using Echarts, associating the formulated follow-up event reasoning rules with the visualized event knowledge graph, predicting the event, and constructing an event graph model.

In an alternative embodiment, modeling the spatial data model according to the spatial environment multi-mode data to obtain a spatial environment data model, including: establishing a resource library according to the chemical equation and the reaction condition, and correlating the chemical equation with the related attribute of the voxel vector model; obtaining the predicted information of the chemical reaction phenomenon and the property change information of the entity when the entity contacts, performing numerical simulation, and converting the chemical reaction formula into a general formula of a chemical model, wherein the general formula of the converted chemical model is as follows:

wherein:m and n are the numbers of reactants and reaction equations in the model respectively,and->Respectively +.>Reaction->Reactant and product coefficients of the components, < >>Is a molecular formula participating in the reaction.

In an alternative embodiment, modeling the spatial data model according to the spatial environment multi-mode data to obtain a spatial environment data model, including: decomposing the space environment data into three layers, wherein the first layer is divided into land, sea, air, sky and net electricity according to the types of the space environment; the second layer decomposes the evaluation index according to the relation between the movable main body and the environmental factors to obtain the evaluation index, wherein the evaluation index at least comprises prediction capability, traffic capability and communication capability; the third layer is a space environment influence factor, and the space environment influence factor at least comprises landform, vegetation, weather, geology and humanity; calculating an impact factor based on the environmental impact factor; establishing a hierarchical model of the system, determining the relation between the factors of the upper layer and the factors of the lower layer, establishing a judgment matrix by adopting a method of comparing the factors of the upper layer and the factors of the lower layer, quantifying the relative importance of each factor in a certain layer, solving the maximum eigenvector of the judgment matrix by using a root method, normalizing and determining the weight of an influence factor; and constructing a complex efficacy model based on the influence factors and the corresponding weights.

In a second aspect, the present application provides a spatial environment data model construction apparatus, including: the digital twin classification module is used for extracting space multi-dimensional element information based on the application scene and carrying out digital twin classification processing on the space multi-dimensional element information to obtain space environment multi-modal data; the model construction module is used for carrying out space data model modeling according to the space environment multi-mode data to obtain a space environment data model; the space environment data model comprises a grid model, a scene model, a voxel vector model, a field model, a flow model, a special effect model, a physical model, a chemical model, a complex efficiency model, a facility model, a relation function model and an event map model; the relation determining module is used for determining the association relation among the data models based on element information, a description method, data types and data organization among the data models in the spatial environment data model; the mapping module is used for mapping the space environment data model to a space environment entity; the space environment entity is in a triplet form formed by an entity structure, attribute elements and association relations.

According to the construction method and the construction device of the space environment data model, the top-down method is adopted, the space environment multi-modal data is obtained based on element information covered by the application scene carding space environment, the space environment multi-modal data is modeled according to the space environment multi-modal data, various types of space environment data models are obtained, and then the relation among the data models is obtained from a plurality of layers of element information, description methods, data types and data organization of each model; various types of data models are mapped to spatial environment entities. The data model constructed by the method covers spatial environment information such as physical objects, phenomena, events and the like of the whole-domain whole-space whole elements, lays a theoretical foundation for the construction of spatial environment entities, and supports the application of twin space.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present application, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flowchart of a method for constructing a spatial environment data model according to an embodiment of the present application;

FIG. 2 is a flowchart of a specific implementation manner according to an embodiment of the present application;

FIG. 3 is a schematic diagram of partial element information of a space environment according to an embodiment of the present application;

FIG. 4 is a schematic diagram of a general framework of spatial environment data according to an embodiment of the present application;

FIG. 5 is a schematic diagram illustrating classification of a data model according to an embodiment of the present application;

FIG. 6 is a schematic diagram of an entity association tuple provided in an embodiment of the present application;

FIG. 7 is a block diagram of an apparatus for constructing a spatial environment data model according to an embodiment of the present application;

fig. 8 is a block diagram of an electronic device according to an embodiment of the present application.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. The components of the embodiments of the present application generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.

Thus, the following detailed description of the embodiments of the application, as presented in the figures, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.

The embodiment of the application provides a method for constructing a spatial environment data model, which is shown in fig. 1, and comprises the following steps:

Step S110, extracting space multi-dimensional element information based on the application scene, and carrying out digital twin classification on the space multi-dimensional element information to obtain space environment multi-modal data.

In one embodiment, the spatial environment multimodal data includes natural environment class data, facility environment class data, social human class data, network electromagnetic class data, and digital twin mechanism model data.

Step S120, modeling a spatial data model according to the spatial environment multi-mode data to obtain a spatial environment data model; the space environment data model comprises a grid model, a scene model, a voxel vector model, a field model, a flow model, a special effect model, a physical model, a chemical model, a complex efficiency model, a facility model, a relation function model and an event map model.

Step S130, determining the association relationship among the data models based on the element information, the description method, the data types and the data organization among the data models in the spatial environment data model.

Step S140, mapping the space environment data model to a space environment entity; the space environment entity is in a triplet form formed by an entity structure, attribute elements and association relations.

For easy understanding, the embodiment of the present application provides a specific implementation manner, and referring to fig. 2, the method includes the following steps S1 to S5:

s1, acquiring 5 kinds of space environment data by adopting a top-down method based on element information covered by the application scene carding space environment.

S2, constructing a space environment data model by adopting a structure mode identification method, and dividing the space environment data model into 12 types of data models to form a space environment data model framework.

S3, designing various data models from 5 aspects of application requirements, basic concepts, description methods, construction methods and data resource construction by adopting a space data model modeling method.

S4, acquiring the relation between the data models from the element information, the description method, the data types and the data organization of the models.

S5, mapping the model to a space environment entity based on concepts and descriptions of various data models.

These five steps are described in detail below.

By adopting a top-down method, each specific application scene in the space environment is firstly analyzed, such as path planning, flood monitoring, rescue material transportation, field reconnaissance, public facility configuration, urban environment dynamic monitoring, equipment maintenance management and the like; and then combing the space environment element information related to each application scene. The spatial environment part element information is shown in fig. 3.

Finally, based on elements such as landforms, vegetation, ocean geography, ocean hydrology and the like contained in the space environment, the data related to the elements are combed according to the professional field according to the classification principle that the space environment data is fully covered and does not cross; the digital twinning concept and connotation are combined, namely, the digital twinning core is to build a mechanism model or a data driving model for the physical entity to describe the structure, the state and the rule of the physical entity, the common data of various mechanism models are researched, and finally 5 kinds of space environment data are obtained.

1) Natural environment class data

The natural environment data comprises geophysical, earth atmosphere, meteorological, terrestrial, marine and cosmic space six-big environment data information.

Geophysics include data such as earth's spin cycle, spin speed, earth's gravitational field, earth's magnetic field, etc. The earth atmosphere includes data such as dry clean atmosphere, water vapor, suspended aerosol particles, and the like. Meteorological includes physical quantities and physical phenomenon data representing atmospheric conditions. Wherein the physical quantity representing the atmospheric state includes air temperature, humidity, wind power, etc.; physical phenomena indicative of atmospheric conditions include wind, clouds, rain, fog, hail, lightning, and the like. Land includes data of land topography, soil, vegetation, land water system, etc. The ocean includes data of sea water temperature, salinity, depth, density, etc. The space comprises solar activity, orbit of the near-earth spacecraft and other data.

2) Facility environment class data

The facility environment class data includes important economic facilities, important traffic facilities, urban important institution facilities, important combat facilities, important logistical equipment facility data, and the like.

Important economic facility data includes oil fields, gas fields, nuclear power stations, thermal power plants, hydroelectric power plants, aircraft manufacturers, and the like. Important traffic facility data include railway military dedicated lines, railroad bridges, highway bridges, tunnels, ports, transportation pipes, and the like. The city important institution facility data comprise important scientific institutions, important higher institutions, comprehensive hospitals and the like. Important combat facility data includes naval bases, command facilities, key plugs, etc.

3) Social personality data

Social humanity data includes border and political region data, educational data, economic data, demographic data, cultural data, religious data, political legal data, ethnic data, and the like. Wherein the boundary and political region data comprises administrative regions, provinces, cities and the like; the education data includes education degree, education expenses, entrance rate, etc.; economic data comprise national production total value, employment rate, income level and the like; population data includes population number, population density, age distribution, gender proportion, population migration, etc.; the cultural data comprises historical relics, cultural tradition and the like; religious data includes religious beliefs, ceremony, and the like; political laws and regulations include laws and regulations, and the like; the ethnic data includes ethnic proportion, ethnic distribution, and the like.

4) Network electromagnetic data

The network electromagnetic data comprises a network and an electromagnetic part. Electromagnetic includes information such as electromagnetic carriers and electromagnetic spectrum. The network contains information such as network carrier, network IP frequency band, etc. The carrier comprises important facilities such as satellite ground stations, television stations, radio stations, communication optical cables, spacecraft launching sites and the like.

5) Digital twin mechanism model data

The mechanism model is a model describing the structure, state and rules of a physical entity based on a mass balance equation, an energy balance equation, a momentum balance equation, certain physical equations, a chemical reaction law, a circuit basic law and the like, and comprises data of chemical reactants, reaction conditions, physical quantities, physical states, fluid density, fluid viscosity and the like. The mechanism model is mainly used for describing an internal mechanism of the entity object in the development and change process, namely a change rule and an implicit relation, and can predict state information of various elements in the space environment at a certain moment.

The idea of the structural pattern recognition is that the pattern is firstly divided into a plurality of sub-patterns; then the sub-pattern is decomposed into simple sub-patterns; finally, decomposing the sub-modes until the research target is met, wherein the simplest sub-mode is called primitive. The application adopts a research method of structure mode recognition based on data such as GB/T13923-2006 basic geographic information element classification and code, GB/T37118-2018 geographic entity space data specification, novel basic mapping system construction test point technical outline, real scene three-dimensional Chinese construction technical outline and the like, establishes a data model easy to recognize and express according to the characteristics of data expression mode, organization form, change rule and the like, and finally generalizes and summarizes the data model to form a space environment data model architecture.

Geological data in the natural environment class are described through a voxel vector model; the data of landform, vegetation, city and the like are described through a scene model or a voxel vector model or a relation function model; cloud, fog, rain, typhoon, air pressure, sea water temperature, sea water salinity and other weather and ocean data are described through a field model or a flow model; the spatial environment data is described by a physical model. The data of important economic facilities, important traffic facilities, important city facilities, important combat facilities and the like in the facility environment class are described by a voxel vector model or a facility model. Boundaries and government areas, ethnic data, economic data, demographic data, cultural data, religious data, etc. are described by a grid model; the carrier data in the network electromagnetic class is described by a voxel vector model or a facility model; the IP band data is described by a physical model. The spatial environment data overall framework is shown in fig. 4.

The method is based on a space data model modeling method, starting from application scenes and application ranges of various models, defining application requirements of various data models, defining basic concepts of the various models, referring to various data standard specifications and various professional field data, defining description methods and construction methods of the various models, and researching the data resource construction methods of the models aiming at updating, storing and application scenes of various data, and finally realizing construction of the various data models.

1) Grid model

Application requirements: and supporting scene applications such as road traffic planning, weapon force deployment, demographics and the like.

Basic concept: the grid model is a data model for describing the outline quantization results of various physical quantities in an abstract domain by uniformly dividing a space into grid cells with different scales by adopting a recursion subdivision method.

The description method comprises the following steps: the mesh model is described by spatial reference mesh, mesh type, mesh subdivision hierarchy, mesh size, mesh coordinate location, mesh address code, cell attributes, etc.

In one embodiment, the modeling of the spatial data model according to the spatial environment multi-mode data to obtain the spatial environment data model may include the following steps 1-1 and 1-2:

step 1-1, marking each level of subdivision grids by adopting 64-bit coding;

step 1-2, selecting a corresponding grid value method based on the type of the grid element attribute to model, and constructing a grid model; the grid element attributes at least comprise population, economy, weather and hydrology, and the value of each grid unit of the grid model is determined by the area attribute value corresponding to the grid center point.

In practical application, the construction method can specifically adopt:

First, 64-bit codes are adopted to identify each level of subdivision grids. The 0 th level mesh grid is identified as 0, and the 1 st level mesh grid codes are 00, 01, 02, and 03, respectively. Starting from the 2 nd stage, the next stage of subdivision grid coding is added with 0, 1, 2 and 3 after the last stage of grid coding.

Then modeling the cell attribute, and adopting an area occupation method for cell attribute data of residential places, government areas and the like; adopting a center attribution method for grid element attribute data of population, economy, weather, hydrology and the like, wherein the value of each grid unit is determined by the area attribute value corresponding to the grid center point; an importance method is adopted for facilities such as electric power, water conservancy and the like with military value and national defense potential.

And (3) data resource construction: the grid model mainly stores data such as population, economy, religion, culture and the like, and is stored by adopting GeoTIFF and JSON files. Based on the changing characteristics of the data, the data is updated periodically, such as once in ten years of population data.

2) Scene model

Application requirements: for reflecting the geospatial distribution morphology across the region.

Basic concept: a scene model is a continuous piece of geographic information data model that reflects the real world geospatial location and morphology over a range of areas.

The description method comprises the following steps: the scene model is described by long radius, flat rate, projection mode, central meridian, banding mode, coordinate unit, ground resolution, medium error, data area, edge condition, etc.

In one embodiment, the modeling of the spatial data model according to the spatial environment multi-mode data to obtain the spatial environment data model may include the following steps 2-1 to 2-6:

step 2-1, constructing a digital elevation model based on digital photogrammetry, laser radar and interferometric radar;

step 2-2, constructing a digital surface model by adopting a regular grid and an irregular grid;

step 2-3, constructing a digital orthographic image by utilizing digital differential correction based on a digital elevation model (Digital Elevation Model, DEM);

step 2-4, constructing a real shot image by utilizing digital differential correction based on a digital surface model (Digital Surface Model, DSM);

step 2-5, generating an oblique photography three-dimensional model through an oblique photography technology;

and 2-6, determining a scene model based on the digital elevation model, the digital surface model, the digital orthographic image, the real shot image and the oblique photography three-dimensional model.

In practical application, the construction method comprises the following steps: constructing a digital elevation model by combining various technologies of digital photogrammetry, laser radar and interference radar; the digital surface model is constructed by adopting two methods, namely a regular grid and an irregular grid; the digital orthophotos are acquired by utilizing a digital differential correction technology based on the DEM; the true shot image is acquired by digital differential correction technology based on DSM; the oblique photography three-dimensional model is generated by oblique photography techniques.

And (3) data resource construction: the digital elevation model, the digital surface model, the digital orthographic image and the real orthographic image are stored by using a GeoTIFF file, and the oblique photography three-dimensional model is stored by using an OSGB file. The scene model data is updated periodically, and the key areas are updated dynamically as required.

3) Voxel vector model

Application requirements: the method provides support for the two-dimensional and three-dimensional expression requirements of the ground feature information (lakes, dams, bridges, houses and the like) of the space environment.

Basic concept: voxel vector models are data models describing natural features, artificial facilities and geographic units that occupy certain and continuous spatial locations in the real world, individually having the same attributes or complete functions.

The description method comprises the following steps: the voxel vector model is described by structure type, morphological content, entity state, LOD hierarchy, geometric information, and the like.

In one embodiment, the modeling of the spatial data model according to the spatial environment multi-mode data to obtain the spatial environment data model may include the following steps 3-1 to 3-3 when implemented:

step 3-1, obtaining real image data, labeling the real image data with a geographic entity, establishing a training sample set, and performing image interpretation on the training sample set to obtain the geographic entity expressed in a two-dimensional form;

Step 3-2, selecting an oblique photography three-dimensional model based on the acquisition precision of the ground object entity, then cutting a triangular net and texture to obtain a three-dimensional geometric shape and texture, and obtaining a three-dimensional expressed geographical entity through preset restoration processing;

and 3-3, constructing a semantic description framework, acquiring semantic information, constructing a hierarchical relationship, and generating a voxel vector model.

In practical application, (1) morphology: including two-dimensional forms and three-dimensional forms. Selecting a proper TDOM as a data source based on the acquisition precision requirement of a basic geographic entity, marking the geographic entity, establishing a training sample set, and obtaining the geographic entity expressed in a two-dimensional form through image interpretation; and selecting a three-dimensional model suitable for oblique photography based on the acquisition precision of the ground object entity, then cutting a triangular net and texture to obtain a three-dimensional geometric shape and texture, and finally obtaining the geographic entity expressed in a three-dimensional form through hole filling, texture repairing and the like.

(2) Attributes: the semantic attribute association is realized by constructing a semantic description framework, acquiring semantic information, carrying out semantic and geometric normalization processing, establishing an aggregation hierarchical relationship and the like.

Build level and data description are shown in the following table:

table 1 build level and data description table

Hierarchy level	Data
		LODa	Point, sketch existing data, automatic construction
LODb	Line, uses DEM and DOM as basic data, uses deep learning algorithm to extract building contour line and elevation information to implement construction
		LOD0	The two-dimensional plane is a two-dimensional geometric primitive formed by the vertical projection of an entity in space
LOD1	Ash box three-dimensional model, constructing three-dimensional graphic primitive according to entity three-dimensional space information
		LOD2	Adding a texture three-dimensional model of a material, and endowing texture properties of the material to a three-dimensional surface on the basis of LOD1 level
LOD3	Adding a three-dimensional model of a real texture, and adding real texture image data of a solid three-dimensional surface on the basis of LOD2 level
		LOD4	Adding a three-dimensional model of fine features of the external contour, measuring the three-dimensional external contour features on the basis of LOD3 level, and carrying out fine expression on each part of the entity
LOD5	Adding a schematic internal structure three-dimensional model, and adding schematic expression to the internal structure of the entity on the basis of LOD4 level
		LOD6	Adding a three-dimensional model of an accurate internal structure, adding accurate internal structure information on the basis of LOD5 level, and acquiring accurate internal three-dimensional coordinate information by adopting a close-range photogrammetry mode
LOD7	Adding an internal structure real texture three-dimensional model, and giving a real texture image to an internal detailed structure according to internal coordinate information on the basis of LOD6 level

And (3) data resource construction: the voxel vector model data is stored by adopting a Shape, geoTIFF file, and a dynamic updating mode is adopted.

4) Facility model

Application requirements: facilities including important economic facilities, important traffic facilities, important city facilities, important combat facilities, etc. are important components of the space environment.

Basic concept: the facility model is a data model for expressing attribute characteristics of important material engineering facilities.

The description method comprises the following steps: the facility model is described by use, material, texture, use state, use environment, and the like.

In one embodiment, the modeling of the spatial data model according to the spatial environment multi-mode data to obtain the spatial environment data model may include the following steps 4-1 and 4-2:

step 4-1, establishing a facility attribute database, and calling facility component geometric structure data based on a voxel vector model;

and 4-2, establishing a corresponding relation between the facility attribute database and the voxel vector model structure to carry out semantic attribute association, acquiring characteristic points of the facility according to the voxel vector model structure, calculating attribute values of the characteristic points, and constructing a facility model.

The construction method comprises the following steps: the construction method is attribute modeling, and the specific modeling steps are as follows: (1) A facility attribute database is established and it is ensured that the data structure of each object class contains unique keywords. (2) Facility component geometry data is invoked using a voxel vector model. (3) And establishing a corresponding relation between the facility attribute database and the voxel vector model structure, realizing semantic attribute association, acquiring characteristic points of the facility according to the voxel vector model structure, calculating attribute values of the characteristic points, and finally realizing construction of a facility model.

And (3) data resource construction: the facility model data is stored by adopting Shape and JSON files, and the update period is not more than one year by adopting a periodical update mode.

5) Field model

Application requirements: the support is provided for the activities such as radar positioning, geomagnetic navigation, weather prediction, ship navigation, lifesaving and salvage and the like. If the ship cannot navigate due to the large fog; the magnetic field information can affect the accuracy of navigation.

Basic concept: the field model regards a geographic phenomenon as a continuous variable or volume, and assigns any position in two-dimensional or three-dimensional space to an attribute value representing the phenomenon.

The description method comprises the following steps: the field model is described by the total geomagnetic intensity, the geomagnetic horizontal component, the declination, the tilt angle, the north component, and the like.

In one embodiment, the field model includes a gravitational field model, a geomagnetic field model, a cloud field model, a wind field model, and a fog field model. The modeling of the spatial data model according to the spatial environment multi-mode data to obtain the spatial environment data model may include the following steps 5-1 to 5-5 when in specific implementation:

step 5-1, calculating a gravity quasi-geoid by a removal-recovery technology, and constructing a gravity field model;

step 5-2, calculating a global model of the geomagnetic field by adopting a spherical harmonic analysis method, and calculating a regional model of the geomagnetic field by adopting a crown harmonic analysis method;

Step 5-3, constructing a cloud field model by calculating three-dimensional cloud cover, characteristic layer cloud cover, cloud bottom height and cloud top height;

5-4, decomposing wind field data into an intrinsic mode and a main coordinate through a characteristic value by adopting an intrinsic orthogonal decomposition method, and constructing a wind field model;

and 5-5, constructing a fog field model by solving the average visibility value in each grid.

The construction method comprises the following steps: (1) gravitational field model: the gravity quasi-geodetic level is calculated using a remove-restore technique. (2) Geomagnetic field model: calculating a global model of the geomagnetic field by adopting a spherical harmonic analysis method; and calculating a regional model of the geomagnetic field by adopting a crown harmonic analysis method. (3) Cloud field model: and constructing a cloud field model by calculating the three-dimensional cloud quantity, the characteristic layer cloud quantity, the cloud bottom height and the cloud top height. (4) Wind field model: and decomposing wind field data into an eigenmode and a main coordinate through a eigenvalue by adopting an eigenorthogonal decomposition method. (5) Mist field model: and constructing a fog field model by solving the average visibility value in each grid.

And (3) data resource construction: the field model data is stored in NetCDF format. The gravity field model update frequency is 10 years/time; the geomagnetic field model is different from month to years according to different spatial scales; the data updating frequency of the cloud field, the fog field and the wind field model Chinese area is 1 time/time, and the global data updating frequency is 3 time/time.

6) Flow model

Application requirements: the analysis and prediction of the relative movement of the fluid can support various command decisions. For example, a river prevents the army from marching at normal temperature, but when the thickness of the frozen river reaches a certain level, the army can pass through.

Basic concept: the flow model is a data model for expressing the time-space change rule of various fluid movements, and comprises a typhoon model, a water flow model and the like.

The description method comprises the following steps: the typhoon flow model is described by element information of typhoons, including time, center position, center air pressure, maximum wind speed, intensity, moving speed, moving direction, and the like. The water flow model is described by measuring time, coordinates of particles at an initial moment, particle flow velocity and acceleration, pressure, density, temperature, surface tension, etc.

The construction method comprises the following steps: typhoon model: initial data such as the maximum wind speed, the movement direction and the like of typhoons are obtained through network resources, a mathematical model is built by combining meteorology and aerodynamics, and the walking path of typhoons is simulated. And (3) a water flow model: and solving the density, speed, pressure and other changes of each particle according to the fluid motion law, and simulating the motion process of the fluid.

And (3) data resource construction: the typhoon model adopts relevant data in a typhoon network of a weather station, is updated every 3 hours, and is stored by adopting a NetCDF file. The water flow model data is stored by adopting PSI files and is dynamically updated as required.

7) Special effect model

Application requirements: the method is used for simulating the effects of explosion, fireworks, clouds, fog, tail gas and the like and providing a realistic virtual environment.

Basic concept: the special effect model is a data model for carrying out virtual simulation on the irregularity of the morphological structure of the physical phenomenon and the variability of the motion state.

The description method comprises the following steps: the special effect model comprises a particle system, a bone animation and dynamic texture mapping. Particle systems are described by the position, direction of motion, speed, acceleration, shape, size, color, texture, transparency, etc. of each particle. Bone animation is described by coordinate system, bone numbering, bone angle, bone displacement, color, length, etc. Dynamic texture mapping is described by texture picture, texture color, texture resolution, texture geometry information, etc.

The construction method comprises the following steps: the special effect model animation construction mainly utilizes three-dimensional modeling software to generate a grid model composed of a series of skeleton nodes, and each node comprises position, rotation, translation and scaling information; searching key frames, and carrying out interpolation calculation on the key frames by adopting a quaternion spherical linear interpolation algorithm, wherein the calculation formula is as follows:

wherein, the liquid crystal display device comprises a liquid crystal display device,is a key frame, p, q are two quaternions needing interpolation, and t is a parameter value between 0 and 1;

The method based on rules is adopted to transform and update the skeleton global, and the skeleton global change formula is as follows:

。

and (3) data resource construction: the particle system model data is stored by adopting a PSI file; the bone animation model data is stored by adopting BVH files; the dynamic texture mapping model data is stored using TIFF files. The special effect data adopts a real-time dynamic updating mode.

8) Physical model

Application requirements: all substances in the space environment have specific structures, properties and motion rules, and the accurate prediction and twin mapping of the changes of the substances in the space environment can be realized by analyzing and calculating based on related physical data.

Basic concept: physical model a data model expressing various physical phenomena and their intrinsic mechanisms within a spatial environment, the classification of which is shown in figure 5.

The description method comprises the following steps: the physical model is described by operator class, influence coefficient, property change, state change, etc.

In one embodiment, the modeling of the spatial data model according to the spatial environment multi-mode data to obtain the spatial environment data model may include the following steps 6-1 and 6-2:

step 6-1, establishing a resource library according to a physical formula, an influence coefficient, a symbol and the like, correlating the physical formula with the related attribute of the voxel vector model,

And 6-2, obtaining physical state change information when the entity contacts, and converting a physical formula into a general formula of a physical model by using a numerical simulation method, wherein the general expression of the physical model is as follows:

where i represents a physical quantity contained in the model, such as velocity, time, course, mass, etc. in the mechanical model,for influencing coefficients, such as dynamic influencing coefficients, friction coefficients, resistance coefficients, thermal conductivity coefficients, etc., y represents the value of the physical model.

And (3) data resource construction: information such as influence coefficients, formulas and the like in the physical model is stored by adopting a JSON file, and is dynamically updated as required.

9) Chemical model

Application requirements: the method is used for space environmental pollution monitoring, chemical toxicity evaluation and the like.

Basic concept: the chemical model is a data model for abstracting chemical problems such as material structure, composition, properties and the like in a space environment.

The description method comprises the following steps: the chemical model is described by information on the nature, state, influence coefficient, property change, state change, etc. of molecules and atoms.

In one embodiment, the modeling of the spatial data model according to the spatial environment multi-mode data to obtain the spatial environment data model may include the following steps 7-1 and 7-2:

Step 7-1, establishing a resource library according to a chemical equation and reaction conditions, and associating the chemical equation with the related attribute of the voxel vector model;

step 7-2, obtaining the predicted information of the chemical reaction phenomenon and the property change information of the entity during the physical contact, performing numerical simulation, and converting the chemical reaction formula into a general formula of a chemical model, wherein the general formula of the converted chemical model is as follows:

In practical application, a resource library is established according to a chemical equation and reaction conditions, the chemical equation is associated with related attributes of a voxel vector model, prediction information of chemical reaction phenomenon and property change information of an entity are obtained when the entity contacts, and finally, a numerical simulation method is utilized to convert a chemical reaction formula into a general formula of the chemical model.

And (3) data resource construction: the chemical model data is stored by adopting a JSON file and is dynamically updated according to the requirement.

10 Event atlas model

Application requirements: support the expression requirement of the state and the operation rule of the event analysis, tracking and backtracking physical world.

Basic concept: the event map model is a data model for dynamically tracking and describing the development and trend of an event according to external conditions and feedback information.

The description method comprises the following steps: event atlas models are typically described in terms of attributes such as event name, event abstract, event type, time, place, object, event impact, event condition, event outcome, etc.

In one embodiment, the modeling of the spatial data model according to the spatial environment multi-mode data to obtain the spatial environment data model may include the following steps 8-1 to 8-4:

step 8-1, carrying out event recognition by adopting a method of jointly recognizing an entity and an event trigger word based on sequence labeling based on network resource data (such as text, video, image, geographic position and the like of an open network), extracting event elements in an event description sentence according to a predefined event representation frame, and judging event types of the event elements, wherein the event types comprise event occurrence types, event influence types and emergency decision types;

step 8-2, constructing candidate event pairs, carrying out standardization processing on structured event elements, describing the similarity of event occurrence events through the similarity of the entity, the event elements and the event semantics, and finally solving the problem of possible event element conflict in the completion process;

Step 8-3, judging event correlation by adopting a method of co-occurrence events, counting occurrence probability among the events with correlation, mapping the event relationship marked in the event corpus to the event relationship, realizing event type relationship discovery, and establishing an event knowledge graph;

and 8-4, visualizing the event knowledge graph by using Echarts, associating the formulated subsequent event reasoning rules with the visualized event knowledge graph, and carrying out event prediction according to the event knowledge graph to construct an event graph model.

In practical application, based on data such as text, video, images, geographic positions and the like of an open network, an event knowledge graph is built by combining an event extraction technology, an information completion technology and a relationship inference technology, and an event association relationship is built, so that an event graph model is built.

And (3) data resource construction: the event map model data is stored by adopting a JSON file and is dynamically updated along with the development of the event.

11 Complex efficacy model

Application requirements: the scenes of forest fire prediction and forecast, flood monitoring, road traffic planning and the like are affected by various factors such as topography, weather and the like, and a complex efficiency model is required to be constructed on the basis of a multi-class data model.

Basic concept: the complex efficacy model is a data model for calculating and evaluating the efficacy of multiple types of monomer data models after cross fusion.

The description method comprises the following steps: the complex performance model is described by a range, a spatial environment influence factor, a weight set, an evaluation index, a weight set, and the like.

In one embodiment, the modeling of the spatial data model according to the spatial environment multi-mode data to obtain the spatial environment data model may include the following steps 9-1 to 9-4:

step 9-1, decomposing the space environment data into three layers, wherein the first layer is divided into land, sea, air, sky and net electricity according to the types of the space environment; the second layer decomposes the evaluation index according to the relation between the movable main body and the environmental factors to obtain the evaluation index, wherein the evaluation index at least comprises prediction capability, traffic capability and communication capability; the third layer is a space environment influence factor, and the space environment influence factor at least comprises landform, vegetation, weather, geology and humane;

step 9-2, calculating an influence factor based on the environmental influence factor;

step 9-3, establishing a system hierarchical level model, determining the relation between the factors of the upper layer and the factors of the lower layer, adopting a method of comparing the factors of the upper layer and the factors of the lower layer to establish a judgment matrix, quantifying the relative importance of each factor in a certain layer, solving the maximum eigenvector of the judgment matrix by using a square root method, normalizing, and determining the weight of the influence factor;

And 9-4, constructing a complex efficiency model based on the influence factors and the corresponding weights.

The construction method comprises the following steps: (1) and decomposing the system. Decomposing the space environment downwards into three layers, wherein the first layer is divided into land, sea, air, sky and net electricity according to the types of the space environment; the second layer decomposes the evaluation index according to the relation between the movable main body and the environmental factors, and can decompose the evaluation index into evaluation indexes such as predictive capacity, traffic capacity, communication capacity and the like; the third layer is a space environment influencing factor such as landform, vegetation, weather, geology, humanity and the like.

(2) And (5) influence factor calculation. Quantitatively analyzing environmental influence factors, and calculating the influence factors through a mathematical formula; and qualitatively analyzing the environmental influence factors, and adopting an evaluation method to enable the environmental influence factors to reach quantification or semi-quantification so as to obtain the numerical values of the influence factors.

(3) And (5) system synthesis. Including determining the impact factor weights and finding the composite evaluation value. The application adopts analytic hierarchy process to determine the weight of the influencing factors. Firstly, establishing a system hierarchical model, and definitely determining the relation between the factors of the upper layer and the factors of the lower layer; then, a judging matrix is established by adopting a pairwise comparison method, and the relative importance of each factor in a certain layer is quantified; and finally solving the maximum eigenvector of the judgment matrix by using a method root and normalizing.

The comprehensive evaluation value characterizes the comprehensive influence of the space environment elements on the activity scene, and is obtained by direct weighted summation.

And (3) data resource construction: the complex efficacy model data is stored by adopting Shape, geoTIFF file and updated dynamically as required.

12 Relational function model

Application requirements: in order to define the relationship between the spatial environment entities and the relationship between the entity attributes, a relationship function model needs to be constructed to describe the entity relationship in a semi-formal manner, so that the dynamic link and the real-time update between the spatial environment entities are facilitated.

Basic concept: the relation function model is a model for representing the relation between entities and the attributes thereof, and can express qualitative and quantitative relation between the entities.

The description method comprises the following steps: the relationship function model is described by entity objects, relationship types, a set of relationship semantic descriptions (near, very near, adjacent), conditional constraints, and the like.

The construction method comprises the following steps: the relationships among the entities include positional relationships, membership relationships, time sequence relationships, geometric constitution relationships, influence relationships, theme relationships, and the like. And acquiring the association relation between the entities by combining the knowledge graph, the deep learning and the rule-based method.

And (3) data resource construction: the relation function model data is stored by adopting a JSON file and synchronously updated along with entity changes.

The scene model and the voxel vector model in the data model are used for describing multi-level space information of an entity and reflecting twin stereotactic; the field model and the flow model are used for expressing meteorological hydrology and marine environment element information; the event map model mainly expresses the evolution process of natural events, artificial events, social events and the like in the social humane environment; the dynamic change mechanism of the description phenomena such as a physical model, a chemical model, a special effect model, a complex efficiency model and the like can reflect the real-time change of a space environment entity, and realize twin dynamics; the relation function model describes the association relation between the entities, and can reflect the association and openness of twinning according to scene expansion.

Firstly, preliminarily obtaining the association relation between the models, namely the structural relation, by analyzing element information of each model; then, along with the gradual definition of the model content, the relationship type among the data models is obtained by analyzing the difference of the description methods of the models, the data types of the models, the data organization and the storage file format of the data organization: collaborative relationship, conversion relationship, bearing relationship, extraction relationship and presentation relationship. The relationship between models is described in table 2 below.

Table 2 data model relationship description table

And analyzing and summarizing the relation between each model and other models according to the relation description table among the models, wherein the relation is shown in the following table 3.

TABLE 3 relationship table between data models

5. The model is mapped to the spatial environment entity based on concepts and descriptions of the various types of data models.

In order to implement the construction of the spatial environment entity, the multi-modal data model needs to be mapped with the spatial environment entity. The application performs normalized description on the space environment entity in the form of the triple of the entity structure, the attribute element and the association relation, and determines the mapping rule of the data model and the space environment entity based on the concept and description of various data models.

The scene model, the voxel vector model and the grid model express the geometrical characteristics of the entity, but the expression mode of each model is different. Wherein the scene model is constructed by taking the physical world as a piece of skin; the voxel vector model is used for dividing the physical world according to the characteristics of the entity object; the mesh model describes the physical geometry through a diagrammatical quantification of various physical quantities within the uniformly rasterized region. Thus, the scene model, voxel vector model, mesh model map to the entity structure tuples.

The field model describes related attribute elements of the geophysical field and the meteorological marine field in the entity; the facility model describes the physical world according to the functional attribute of the entity object; the flow model is used for describing fluid motion related attributes of the entity; the physical model describes physical change rules of entities such as sound, light, force, heat, magnetism and the like; the chemical model describes the chemical change rule of the entity; the complex efficacy model expresses holographic information elements possibly carried by an entity; the special effect model describes the nature phenomenon motion state attributes. Thus, field models, facility models, flow models, physical models, chemical models, complex efficacy models, effect models map to entity property tuples.

The event map model mainly expresses causal relationship, time sequence relationship, cis-bearing relationship and the like among events; the relation function model mainly expresses azimuth relation, upper and lower relation, command relation, influence relation and the like among the entities. Thus, the event map model and the relation function model are mapped to the entity association relation tuple, as shown in fig. 6.

In conclusion, the application constructs a complete set of space environment data model architecture. The system covers space environment information such as physical objects, phenomena, events and the like of all elements in the whole space of the whole domain, has the remarkable characteristics of materialization, semanteme, structuring, multidimensional dynamics, the whole space and the like, and supports analysis decisions of the multi-domain space such as geographic space, socioeconomic space and the like. The application defines the relation among the models in the spatial environment data model, is beneficial to increasing the reusability of the model and improving the quality of the data model. The application defines the mapping rule of various data models and space environment entities, lays a theoretical foundation for the construction of the space environment entities, and can support the application of twin space.

For the above method for constructing a spatial environment data model, an embodiment of the present application provides a device for constructing a spatial environment data model, as shown in fig. 7, where the device includes the following parts:

the digital twin classification module 710 is configured to extract spatial multi-dimensional element information based on an application scene, and perform digital twin classification on the spatial multi-dimensional element information to obtain spatial environment multi-modal data;

the model construction module 720 is configured to perform spatial data model modeling according to the spatial environment multi-mode data to obtain a spatial environment data model; the space environment data model comprises a grid model, a scene model, a voxel vector model, a field model, a flow model, a special effect model, a physical model, a chemical model, a complex efficiency model, a facility model, a relation function model and an event map model;

a relationship determining module 730, configured to determine an association relationship between each data model based on element information, a description method, a data type and a data organization between each data model in the spatial environment data model;

a mapping module 740 for mapping the spatial environment data model to a spatial environment entity; the space environment entity is in a triplet form formed by an entity structure, attribute elements and association relations.

In an alternative embodiment, the model building module 720 is further configured to: aiming at the grid model, marking each level of subdivision grids by adopting 64-bit coding; selecting a corresponding grid value method based on the type of the grid element attribute to carry out modeling, and constructing a grid model; the grid element attributes at least comprise population, economy, weather and hydrology, and the value of each grid unit of the grid model is determined by the area attribute value corresponding to the grid center point.

In an alternative embodiment, the model building module 720 is further configured to: constructing a digital elevation model based on digital photogrammetry, lidar and interferometric radar; constructing a digital surface model by adopting a regular grid and an irregular grid; constructing a digital orthographic image by utilizing digital differential correction based on the DEM; constructing a true shot image based on the DSM by utilizing digital differential correction; generating a oblique photography three-dimensional model by an oblique photography technology; the scene model is determined based on a digital elevation model, a digital surface model, a digital orthographic image, a true orthographic image, and a oblique photography three-dimensional model.

In an alternative embodiment, the model building module 720 is further configured to: obtaining real shooting image data, marking the real shooting image data with a geographic entity, establishing a training sample set, and performing image interpretation on the training sample set to obtain the geographic entity expressed in a two-dimensional form; selecting an oblique photography three-dimensional model based on the acquisition precision of the ground object entity, then cutting a triangular net and texture to obtain a three-dimensional geometric shape and texture, and obtaining a three-dimensional form expressed geographical entity through preset restoration processing; and constructing a semantic description framework, acquiring semantic information, constructing a hierarchical relationship, and generating a voxel vector model.

In an alternative embodiment, the model building module 720 is further configured to: establishing a facility attribute database, and calling facility component geometric structure data based on a voxel vector model; establishing a corresponding relation between a facility attribute database and a voxel vector model structure to carry out semantic attribute association, acquiring characteristic points of facilities according to the voxel vector model structure, calculating attribute values of the characteristic points, and constructing a facility model.

In an alternative embodiment, the field model includes a gravitational field model, a geomagnetic field model, a cloud field model, a wind field model, and a fog field model; the model building module 720 is further configured to: calculating a gravity quasi-geoid by a removal-recovery technology, and constructing a gravity field model; calculating a global model of the geomagnetic field by adopting a spherical harmonic analysis method, and calculating a regional model of the geomagnetic field by adopting a crown harmonic analysis method; a cloud field model is built by calculating three-dimensional cloud cover, characteristic layer cloud cover, cloud bottom height and cloud top height; decomposing wind field data into an intrinsic mode and a main coordinate through a characteristic value by adopting an intrinsic orthogonal decomposition method, and constructing a wind field model; and constructing a fog field model by solving the average visibility value in each grid.

In an alternative embodiment, the model building module 720 is further configured to: establishing a resource library according to a physical formula, influence coefficients, symbols and the like; correlating a physical formula with the related attribute of the voxel vector model to acquire physical state change information when the entity contacts; the physical formula is converted into a general formula of a physical model by using a numerical simulation method, and the physical model has the following unified expression:

In an alternative embodiment, the model building module 720 is further configured to: based on network resource data, carrying out event recognition by adopting a method of jointly recognizing an entity and an event trigger word based on sequence labeling, extracting event elements in an event description sentence according to a predefined event representation frame, and judging event types of the event elements, wherein the event types at least comprise event occurrence types, event influence types and emergency decision types; constructing candidate event pairs, carrying out standardization processing on structured event elements, describing the similarity of event types through the similarity of the three aspects of entities, the event elements and the event semantics, and solving the problem of event element conflict possibly generated in the completion process; judging event correlation by adopting a co-occurrence event method, counting occurrence probability among events with correlation, mapping the marked relationship in an event corpus to event relationship, and establishing an event knowledge graph; and visualizing the event knowledge graph by using Echarts, associating the formulated follow-up event reasoning rules with the visualized event knowledge graph, predicting the event, and constructing an event graph model.

In an alternative embodiment, the model building module 720 is further configured to: establishing a resource library according to the chemical equation and the reaction condition, and correlating the chemical equation with the related attribute of the voxel vector model; obtaining the predicted information of the chemical reaction phenomenon and the property change information of the entity when the entity contacts, performing numerical simulation, and converting the chemical reaction formula into a general formula of a chemical model, wherein the general formula of the converted chemical model is as follows:

In an alternative embodiment, the model building module 720 is further configured to: decomposing the space environment data into three layers, wherein the first layer is divided into land, sea, air, sky and net electricity according to the types of the space environment; the second layer decomposes the evaluation index according to the relation between the movable main body and the environmental factors to obtain the evaluation index, wherein the evaluation index at least comprises prediction capability, traffic capability and communication capability; the third layer is a space environment influence factor, and the space environment influence factor at least comprises landform, vegetation, weather, geology and humane; calculating an impact factor based on the environmental impact factor; establishing a hierarchical model of the system, determining the relation between the factors of the upper layer and the factors of the lower layer, establishing a judgment matrix by adopting a method of comparing the factors of the upper layer and the factors of the lower layer, quantifying the relative importance of each factor in a certain layer, solving the maximum eigenvector of the judgment matrix by using a root method, normalizing and determining the weight of an influence factor; and constructing a complex efficacy model based on the influence factors and the corresponding weights.

The embodiment of the present application further provides an electronic device, as shown in fig. 8, which is a schematic structural diagram of the electronic device, where the electronic device 100 includes a processor 81 and a memory 80, where the memory 80 stores computer executable instructions that can be executed by the processor 81, and the processor 81 executes the computer executable instructions to implement any one of the above spatial environment data model building methods.

In the embodiment shown in fig. 8, the electronic device further comprises a bus 82 and a communication interface 83, wherein the processor 81, the communication interface 83 and the memory 80 are connected by the bus 82.

The memory 80 may include a high-speed random access memory (RAM, random Access Memory), and may further include a non-volatile memory (non-volatile memory), such as at least one magnetic disk memory. The communication connection between the system network element and at least one other network element is implemented via at least one communication interface 83 (which may be wired or wireless), and may use the internet, a wide area network, a local network, a metropolitan area network, etc. Bus 82 may be an ISA (Industry Standard Architecture ) bus, a PCI (Peripheral Component Interconnect, peripheral component interconnect standard) bus, or EISA (Extended Industry Standard Architecture ) bus, among others. The bus 82 may be classified as an address bus, a data bus, a control bus, or the like. For ease of illustration, only one bi-directional arrow is shown in FIG. 8, but not only one bus or type of bus.

The processor 81 may be an integrated circuit chip with signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in the processor 81 or by instructions in the form of software. The processor 81 may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU for short), a network processor (Network Processor, NP for short), etc.; but also digital signal processors (Digital Signal Processor, DSP for short), application specific integrated circuits (Application Specific Integrated Circuit, ASIC for short), field-programmable gate arrays (Field-Programmable Gate Array, FPGA for short) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present application may be embodied directly in the execution of a hardware decoding processor, or in the execution of a combination of hardware and software modules in a decoding processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in a memory, and the processor 81 reads the information in the memory, and in combination with its hardware, completes the steps of the spatial environment data model construction method of the foregoing embodiment.

The embodiment of the application also provides a computer readable storage medium, which stores computer executable instructions that, when being called and executed by a processor, cause the processor to implement the above-mentioned spatial environment data model construction method, and the specific implementation can be found in the foregoing method embodiments, which are not repeated herein.

The computer program product of the method for constructing a spatial environment data model provided by the embodiment of the present application includes a computer readable storage medium storing program codes, where the instructions included in the program codes may be used to execute the method described in the foregoing method embodiment, and specific implementation may refer to the method embodiment and will not be described herein.

The relative steps, numerical expressions and numerical values of the components and steps set forth in these embodiments do not limit the scope of the present application unless it is specifically stated otherwise.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer readable storage medium executable by a processor. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

In the description of the present application, it should also be noted that, unless explicitly specified and limited otherwise, the terms "disposed," "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present application will be understood in specific cases by those of ordinary skill in the art.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the application.

Claims

1. The method for constructing the spatial environment data model is characterized by comprising the following steps of:

Extracting space multi-dimensional element information based on an application scene, and performing digital twin classification processing on the space multi-dimensional element information to obtain space environment multi-modal data;

modeling a spatial data model according to the spatial environment multi-mode data to obtain a spatial environment data model; the space environment data model comprises a grid model, a scene model, a voxel vector model, a field model, a flow model, a special effect model, a physical model, a chemical model, a complex efficiency model, a facility model, a relation function model and an event map model;

determining the association relation among the data models based on element information, a description method, data types and data organization among the data models in the spatial environment data model;

mapping the spatial environment data model to a spatial environment entity; the space environment entity is in a triplet form formed by an entity structure, attribute elements and association relations.

2. The method for constructing a spatial environment data model according to claim 1, wherein the modeling of the spatial environment data model according to the spatial environment multi-modal data to obtain the spatial environment data model comprises:

Identifying each level of subdivision grids by adopting 64-bit coding;

selecting a corresponding grid value method based on the type of the grid element attribute to carry out modeling, and constructing a grid model; the grid element attribute at least comprises population, economy, weather and hydrology, and the value of each grid unit of the grid model is determined by the area attribute value corresponding to the grid center point.

3. The method for constructing a spatial environment data model according to claim 1, wherein the modeling of the spatial environment data model according to the spatial environment multi-modal data to obtain the spatial environment data model comprises:

obtaining real projective image data, labeling the real projective image data with a geographic entity, establishing a training sample set, and performing image interpretation on the training sample set to obtain the geographic entity expressed in a two-dimensional form;

selecting an oblique photography three-dimensional model based on the acquisition precision of the ground object entity, then cutting a triangular net and texture to obtain a three-dimensional geometric shape and texture, and obtaining a three-dimensional form expressed geographical entity through preset restoration processing;

and constructing a semantic description framework, acquiring semantic information, constructing a hierarchical relationship, and generating a voxel vector model.

4. A method of constructing a spatial environment data model according to claim 1 or 3, wherein the modeling of the spatial environment data model based on the spatial environment multi-modal data to obtain the spatial environment data model comprises:

Establishing a facility attribute database, and calling facility component geometric structure data based on the voxel vector model;

establishing a corresponding relation between a facility attribute database and a voxel vector model structure to carry out semantic attribute association, acquiring characteristic points of facilities according to the voxel vector model structure, calculating attribute values of the characteristic points, and constructing a facility model.

5. The method for constructing a spatial environment data model according to claim 1, wherein the field model includes a gravitational field model, a geomagnetic field model, a cloud field model, a wind field model, and a fog field model;

modeling the spatial data model according to the spatial environment multi-mode data to obtain a spatial environment data model, including:

calculating a gravity quasi-geoid by a removal-recovery technology, and constructing a gravity field model;

calculating a global model of the geomagnetic field by adopting a spherical harmonic analysis method, and calculating a regional model of the geomagnetic field by adopting a crown harmonic analysis method;

a cloud field model is built by calculating three-dimensional cloud cover, characteristic layer cloud cover, cloud bottom height and cloud top height;

decomposing wind field data into an intrinsic mode and a main coordinate through a characteristic value by adopting an intrinsic orthogonal decomposition method, and constructing a wind field model;

and constructing a fog field model by solving the average visibility value in each grid.

6. The method for constructing a model of spatial environment data according to claim 1, wherein,

establishing a resource library according to a physical formula, an influence coefficient and a symbol aiming at the physical model, and associating the physical formula with the related attribute of the voxel vector model;

the physical state change information during physical contact is obtained, a physical formula is converted into a general formula of a physical model by using a numerical simulation method, and the physical model has the following unified expression:

wherein i represents a physical quantity contained in the model, said physical quantity comprising at least a velocity, a time, a distance, a mass in the mechanical model,for the influence coefficient, the influence coefficient at least comprises a dynamic influence coefficient, a friction coefficient, a resistance coefficient and a heat conduction coefficient, and y represents the value of the physical model.

7. The method for constructing a model of spatial environment data according to claim 1, wherein,

based on network resource data, carrying out event recognition by adopting a method of jointly recognizing an entity and an event trigger word based on sequence labeling, extracting event elements in event description sentences according to a predefined event representation frame, and judging event types of the event elements, wherein the event types at least comprise event occurrence types, event influence types and emergency decision types;

Constructing candidate event pairs, carrying out standardization processing on structured event elements, and describing the similarity of event types through the similarity of the three aspects of entities, the event elements and the event semantics;

judging event correlation by adopting a co-occurrence event method, counting occurrence probability among events with correlation, mapping the marked relationship in an event corpus to event relationship, and establishing an event knowledge graph;

and visualizing the event knowledge graph by using Echart, associating the formulated follow-up event reasoning rule with the visualized event knowledge graph, predicting the event, and constructing an event graph model.

8. The method for constructing a spatial environment data model according to claim 1, wherein the modeling of the spatial environment data model according to the spatial environment multi-modal data to obtain the spatial environment data model comprises:

establishing a resource library according to the chemical equation and the reaction condition, and correlating the chemical equation with the related attribute of the voxel vector model;

obtaining the predicted information of the chemical reaction phenomenon and the property change information of the entity when the entity contacts, performing numerical simulation, and converting the chemical reaction formula into a general formula of a chemical model, wherein the general formula of the converted chemical model is as follows:

Wherein:m and n are the numbers of reactants and equations in the model, respectively, +.>Andrespectively +.>Reaction->Reactant and product coefficients of the components, < >>Is a molecular formula participating in the reaction.

9. The method for constructing a spatial environment data model according to claim 1, wherein the modeling of the spatial environment data model according to the spatial environment multi-modal data to obtain the spatial environment data model comprises:

decomposing the space environment data into three layers, wherein the first layer is divided into land, sea, air, sky and net electricity according to the types of the space environment; the second layer decomposes the evaluation index according to the relation between the movable main body and the environmental factors to obtain the evaluation index, wherein the evaluation index at least comprises prediction capability, traffic capability and communication capability; the third layer is a space environment influence factor, and the space environment influence factor at least comprises landform, vegetation, weather, geology and humanity;

calculating an impact factor based on the environmental impact factor;

establishing a hierarchical model of the system, determining the relation between the factors of the upper layer and the factors of the lower layer, establishing a judgment matrix by adopting a method of comparing the factors of the upper layer and the factors of the lower layer, quantifying the relative importance of each factor in a certain layer, solving the maximum eigenvector of the judgment matrix by using a root method, normalizing and determining the weight of an influence factor;

And constructing a complex efficacy model based on the influence factors and the corresponding weights.

10. A spatial environment data model construction apparatus, comprising:

the digital twin classification module is used for extracting space multi-dimensional element information based on an application scene, and carrying out digital twin classification processing on the space multi-dimensional element information to obtain space environment multi-modal data;

the model construction module is used for carrying out space data model modeling according to the space environment multi-mode data to obtain a space environment data model; the space environment data model comprises a grid model, a scene model, a voxel vector model, a field model, a flow model, a special effect model, a physical model, a chemical model, a complex efficiency model, a facility model, a relation function model and an event map model;

the relation determining module is used for determining the association relation among the data models based on element information, a description method, data types and data organization among the data models in the spatial environment data model;

the mapping module is used for mapping the space environment data model to a space environment entity; the space environment entity is in a triplet form formed by an entity structure, attribute elements and association relations.