CN112613209A

CN112613209A - Full hexahedron unit encryption method based on finite element model

Info

Publication number: CN112613209A
Application number: CN202011497634.4A
Authority: CN
Inventors: 庄三少; 陈秀华
Original assignee: Shanghai Jiaotong University
Current assignee: Shanghai Jiaotong University
Priority date: 2020-12-17
Filing date: 2020-12-17
Publication date: 2021-04-06

Abstract

The invention relates to a full hexahedron unit encryption method based on a finite element model, which carries out topological data reconstruction on hexahedron units based on the finite element model; constructing topological unit structure data of the converted units; dividing the topological unit structure data into a target unit and peripheral units, wherein the target unit is a unit needing to be subdivided, expressing each unit by adopting unit topological structure data, and acquiring a transition unit of a hexahedral unit; and finally, updating the topological data of the unit model, and repeatedly carrying out multi-stage subdivision. Compared with the prior art, the method has the advantages of capability of realizing the subdivision of the hexahedron element finite element model at any position and in any scale, automation, flexibility in realization and the like.

Description

Full hexahedron unit encryption method based on finite element model

Technical Field

The invention relates to the technical field of finite element modeling, in particular to a full hexahedron unit encryption method based on a finite element model.

Background

In the field of fatigue crack propagation analysis, the unit density and the unit type of the crack tip have great influence on the calculation accuracy of the stress intensity factor. In general, the tetrahedral unit mesh can approximately represent geometric bodies with any shapes, and is widely applied in the field of structural finite element analysis. However, the tetrahedral units cannot calculate the shear stress, and compared with the hexahedral units, the number of the tetrahedral units is several times larger than that of the hexahedral units under the condition of obtaining the same calculation accuracy. And the hexahedron unit can introduce an additional shape function, so that the calculation accuracy can be further improved. Compared with a tetrahedral unit, the hexahedral unit is widely used in the field of fatigue crack propagation finite element analysis due to high calculation precision.

However, compared with the tetrahedral unit, the hexahedral unit is difficult to divide, and cannot be automatically divided for a complex model. The finite element analysis calculation precision is related to the element density, the element density is generally required to be high in some key areas, and sparse elements are used at some far ends, so that satisfactory calculation precision can be obtained under the condition that the number of the elements is not large, and therefore the element local encryption is very important. The tetrahedral units can be partially encrypted in general, the current unit division algorithm and unit partial encryption algorithm of the tetrahedral units are mature, automatic unit encryption can be performed, but the calculation accuracy of the tetrahedral units is poorer than that of the hexahedral units, and in order to obtain better calculation accuracy, finite element calculation is generally recommended to be performed by using the hexahedral units.

The hexahedron unit finite element model can be generated by adopting common finite element preprocessing software, but one problem commonly existing at present is how to partially encrypt the hexahedron unit. The general method is to specify a very fine unit size when the finite element modeling unit is generated, so that the purpose of unit subdivision can be achieved, but the model units are increased in multiples, and encryption is also performed in some areas which do not need encryption, so that the method is obviously not suitable, the processing cost is increased, and the method is not flexible. Generally, the cost for regenerating the finite element model is too large, most of the existing models are encrypted, and how to partially encrypt the existing models is obvious that the existing finite element preprocessing software cannot partially encrypt hexahedron units.

At present, for a unit local encryption method of a hexahedral unit, some transition template methods are proposed in the prior art, and the methods provide a certain technical basis for realizing an automatic encryption method of the hexahedral unit. The Chinese invention patent CN103729506A provides a complex model hexahedron unit remodeling method, but the method can only encrypt the swept units, is essentially a two-dimensional surface unit encryption method, and cannot be applied to the encryption of three-dimensional hexahedron local units with any structures, and the method is only used when generating finite element units, cannot operate the existing finite element model, and needs a tetrahedron and a pentahedron for temporary transition in a transition area, needs a large amount of manual operation, and cannot be performed automatically. Chinese patent CN103116682A proposes a method for constructing crack elements of finite element holes, which can perform element generation and encryption on hole regions, but is limited to generating tetrahedral elements. In summary, no method for local multilevel encryption of the full hexahedron unit automation unit is proposed at present. Obviously, the method for encrypting the local area of any hexahedral unit becomes a technical bottleneck in the industry at present, and a corresponding technical method is urgently needed to meet the needs in the industry.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provide a full hexahedron unit encryption method based on a finite element model, which applies a topological geometry data structure method to finite element details and can carry out optional local multiple automatic full hexahedron unit encryption on the full hexahedron finite element model.

The purpose of the invention can be realized by the following technical scheme:

a full hexahedron unit encryption method based on finite element model is based on topological geometry data structure, and includes the following steps:

s1: carrying out topological data conversion on the finite element hexahedron unit, namely reconstructing topological data of the hexahedron unit, and specifically comprising the following steps:

1.1) defining each hexahedral cell as quad8, whose topology consists of 6 topological faces;

1.2) defining each topological surface in the unit as face, wherein each topological surface consists of 4 topological edges;

1.3) defining each topological edge as edge, which is composed of two topological points;

1.4) defining each topological point as vertex, wherein the topological point is the minimum topological element and represents a space node position and has a unique id;

1.5) the topology elements are shared with each other, i.e. two adjacent cells share a topology surface.

After the reconstruction steps are implemented, the topological unit model data includes: and a group of shared nodes, edges and faces, and the shared topological elements are combined into unit data.

S2: the set of cells is divided into a target cell and remaining cells and represented by a topological cell data structure.

This operation is performed based on the topology unit data structure. The unit description data file comprises node coordinate information and unit node information, and unit topological structure information can be constructed by reading the file. The specific content of the unit description data file comprises a node id, a node coordinate, a unit id and an id list consisting of unit nodes, and the unit data can be generated into unit topological structure data through unit data analysis.

The specific contents of the generated unit topology structure data are as follows:

(a1) firstly, establishing unit node topology data vertex, establishing a unit topology data structure vertex according to the read id and coordinate values, and storing the unit topology data structure vertex in a global database;

(a2) unit topology data creation is carried out, and a unit id and 8 node ids quoted by the unit are obtained through file analysis;

(a3) creating a topology surface according to 8 node ids, wherein one unit consists of 6 surfaces, the nodes are orderly grouped to obtain 6 groups of topology surface node groups, and each node group consists of 4 node ids;

(a4) and constructing the topology surface of the 4 node groups, calling a topology surface construction algorithm to construct the topology surface, and completing the construction of the topology unit after completing the construction of the 4 topology surfaces.

Starting from inputting four topology node ids, the topology surface construction algorithm specifically comprises the following steps:

(b1) searching a topology node vertex according to the four input topology node ids to obtain 4 topology nodes;

(b2) one topological surface consists of 4 topological edges which are connected end to end, each topological edge consists of 2 topological nodes, the 4 topological nodes are grouped to obtain 4 topological edge input data, and each topological edge input data consists of 2 topological nodes;

(b3) and performing topology edge creation on each topology edge input data, and completing the creation of the topology edge.

Starting from inputting 2 topology node ids, the specific steps of the topology edge creation algorithm comprise:

(c1) searching for a topological edge according to input topological node data;

(c2) traversing the data structure of the topological edges, and searching whether the topological edges meet the equivalence (namely the id is the same) of the topological nodes and the input nodes;

(c3) if the topological edge described in the step (c2) exists, directly returning the reference of the edge (namely, sharing the topological edge);

(c4) if the topology edge described in (c2) does not exist, creating a new topology edge according to the two input nodes;

(c5) and completing the creation of the topological edge.

S3: and implementing a hexahedral cell transition strategy based on the cell topology data.

The topological cells are first classified into two types: the device comprises a target unit and peripheral units, wherein the target unit is a unit needing to be subdivided, and the peripheral units are units not needing to be subdivided. The target unit and the peripheral unit are marked correspondingly in advance.

From the view point of the topological surface, the topological surface is composed of three states: completely in the target cell, completely in the peripheral cell and in between, the three states are described as follows: a target unit surface, a peripheral unit surface and a transition unit surface;

from the view of the target surface state contained in the topological unit, there are several possibilities: full inclusion target cell plane S0, mixed state S1, mixed state S2, and full inclusion peripheral cell plane S3. All the faces including the target unit face S0 are peripheral unit faces, one face in the mixed state S1 is a target unit face, the remaining faces are peripheral unit faces, two adjacent faces in the mixed state S2 are target unit faces, the remaining faces are peripheral unit faces, and all the faces including the peripheral unit face S3 are target unit faces.

A topology edge belongs to one or more topology surfaces, and the topology edge can be divided into three conditions according to the different types of the topology surfaces where the topology edge is located: do not contain target cell plane E1 and contain cell plane E2. The subdivision process of the topological edges of different states is different.

The topological surface can be divided into five states of G1, G2, G3, G4 and G5, wherein the G1 state comprises 4 topological edges of E1 states; the G2 states include a topology ratio edge of 3E 1 states and 1E 2 states; the G3 states include a topology ratio edge of 2E 1 states and 2E 2 states; the G4 states include 1 topological edge of the E1 state and 3 topological ratio edges of the E2 state; the G5 state includes topological specific edges of 4E 2 states, where the G3 and G4 states are not present. And the topological surfaces in different states are subdivided according to different subdivision strategies.

The unit transition carries out unit subdivision based on the topological data structure, namely carrying out subdivision operation on topological elements. And the subdivision starts from the minimum element, firstly carries out topology edge subdivision, then carries out topology surface subdivision, and then carries out topology unit subdivision. Namely, the subdivision sequence is topological edge subdivision, topological surface subdivision and topological unit subdivision.

S4: and updating the topological data of the unit model, and repeatedly performing multi-stage subdivision to increase the unit density and finish encryption processing.

The updating of cell topology data is used for multi-level subdivision, and the steps comprise:

4.1) storing the new topological structure generated after the unit in a new hierarchical database, wherein the new topological structure comprises a newly generated topology edge, a topology surface and a topological unit;

4.2) the newly generated nodes are uniformly stored in a node database, and the node topological structure database is globally shared;

4.3) automatically identifying the peripheral unit and the target unit;

in the invention, an initial unit (a topological unit before subdivision operation) is marked in advance, namely, a target unit and a peripheral unit are respectively marked correspondingly; the subdivision operation is to subdivide the target unit, and the subdivided new unit inherits the mark of the target unit, namely the subdivided units are all the target units.

And automatically identifying the peripheral unit and the target unit based on the marks in the following specific modes:

for each subdivided target unit, traversing all the surfaces of the subdivided target unit, searching a shared surface between the target unit and the peripheral unit, converting the mark of the unit where the surface is located from the target unit to the mark of the peripheral unit after the surface is found, and further completing identification.

4.4) subdividing again on the newly generated hierarchical database;

and designating a cell size, comparing the lateral size of each subdivided cell in the newly generated hierarchical database with the cell size, and if the subdivided lateral size is larger than the size of the designated cell, subdividing the cell again according to the subdividing operation.

4.5) go to step 4.1) to perform multi-level subdivision and increase the cell density again, thus realizing more effective encryption processing.

Compared with the prior art, the full hexahedron unit encryption method based on the finite element model at least has the following beneficial effects:

the method comprises the steps that firstly, a data structure of a hexahedral unit is described by adopting a topological geometric data structure method, automatic unit subdivision and multi-level subdivision can be performed on the basis of the data structure, unit subdivision of any scale at any position of a finite element model of the hexahedral unit can be realized, unit local refinement can be realized under the condition that the number of units is not increased, and solving precision is improved;

secondly, the transition of a full hexahedron unit can be met, a tetrahedron or pentahedron unit does not need to be introduced, manual operation is not needed, and the whole process is carried out automatically;

thirdly, the method can be realized by secondary development of the existing finite element pretreatment software, and can also be realized by independent development, and the realization method is very flexible.

Drawings

FIG. 1 is a schematic diagram of a process of converting topological data of a finite element hexahedral element in an embodiment;

FIG. 2 is a diagram of an embodiment of a topology unit sharing topology surface;

FIG. 3 is a diagram illustrating a process of parsing a data file and creating a topology element according to an embodiment;

FIG. 4 is a schematic diagram of an embodiment of creating a topology face and a topology edge;

FIG. 5 is a schematic diagram of an embodiment of a topology classification;

FIG. 6 is a diagram illustrating the classification of topology units in an embodiment;

FIG. 7 is a schematic illustration of face and edge subdivision in an embodiment;

FIG. 8 is a schematic view of an exemplary hexahedral cell transition template;

FIG. 9 is a schematic diagram of an embodiment of multi-stage subdivision;

FIG. 10 is a schematic diagram of multi-stage subdivision in the example.

Detailed Description

The invention is described in detail below with reference to the figures and specific embodiments. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, shall fall within the scope of protection of the present invention.

Examples

The invention relates to a full hexahedron unit encryption method based on a finite element model, which comprises the following steps:

step one, topological data conversion of the finite element hexahedron unit.

And step two, dividing the unit set into a target unit and a residual unit, and representing the target unit and the residual unit by using a topological unit data structure.

And thirdly, implementing a hexahedral unit transition strategy based on the unit topology data.

And step four, updating the topological data of the unit model, and repeatedly carrying out multi-stage subdivision.

Fig. 1 is a schematic diagram of a finite element hexahedral element topology data conversion process according to an embodiment of the present invention. Fig. 1 introduces the topology data structure in one cell and if the topology data structure is applied to hexahedral cells. The cell is represented by quad8, the cell data includes quad8 ═ { f1, f2, f3, f4, f5, f6}, i.e. quad8 cell has six faces f1, f2, f3, f3, f4, f5 and f 6: f1 ═ e1, e9, e2, e6}, f2 ═ e2, e10, e3, e7}, f3 ═ e3, e11, e4, e8}, f4 ═ e5, e6, e7, e8}, f5 ═ e9, e10, e11, e12}, f6 ═ e1, e12, e4, e5 }.

Each surface consists of four edges respectively, wherein different surfaces share the same edge; the composition of edge is as follows: e1 ═ v1, v2}, e2 ═ v3, v4}, e3 ═ v7, v8}, e4 ═ v5, v6}, e5 ═ v1, v5}, e6 ═ v1, v4}, e7 ═ v4, v8}, e8 ═ v5, v8}, e9 ═ v2, v3}, e10 ═ v3, v7}, e11 ═ v7, v12}, e12 ═ v2, v6 }. Each edge consists of two vertexs, different edges sharing the same vertex. vertex is the smallest topological element, and a quad8 is composed of 8 vertexs, and a complete topological relation is formed by edge and face.

Fig. 2 is a schematic diagram of a topology unit sharing topology surface according to an embodiment of the present invention. FIG. 2 depicts the topology sharing relationship with two topology cells A and B, i.e., cell A and cell B share a topology surface F1. Similarly, if a plurality of units are connected with each other, a plurality of unit sharing surfaces appear; similarly, multiple faces may share the same edge; multiple edges may also share the same node.

Fig. 3 is a schematic diagram of the reading unit data file parsing and topology element creation process of the present invention. Starting from the analysis of the input unit data, the specific steps are described as follows:

1. firstly, establishing unit node topology data vertex, establishing a unit topology data structure vertex according to the read id and coordinate values, and storing the unit topology data structure vertex in a global database;

2. then, unit topology data is created, and a unit id and 8 node ids quoted by the unit are obtained through file analysis;

3. the creation of a topology surface can be carried out according to 8 node ids, one unit consists of 6 surfaces, the nodes are orderly grouped to obtain 6 groups of topology surface node groups, and each node group consists of 4 node ids;

4. and constructing the topology surfaces of the 4 node groups, calling a topology surface construction algorithm (figure 4) to construct the topology surfaces, and completing the construction of the topology units after completing the construction of the 4 topology surfaces.

Fig. 4 is a schematic diagram of creating a topology surface and a topology edge according to an embodiment of the present invention. Starting from inputting four topology node ids, the specific steps are described as follows:

1. searching a topology node vertex according to the four input topology node ids to obtain 4 topology nodes;

2. one topological surface consists of 4 topological edges which are connected end to end, each topological edge consists of 2 topological nodes, the 4 topological nodes are grouped to obtain 4 topological edge input data, and each topological edge input data consists of 2 topological nodes;

3. and (3) performing topology edge creation on each topology edge input data, wherein a creation algorithm is shown in FIG. 4, and the creation of the topology surface is completed after the creation of the topology edge is completed.

Fig. 5 is a schematic diagram of topology classification provided in the embodiment of the present invention. Topological cells are divided into two types: a target cell and a peripheral cell, wherein the target cell needs to be subdivided and the peripheral cell is a cell that does not need to be subdivided. Usually, the finite element elements cannot be divided into very thin parts due to the consideration of computing power, otherwise, the computation amount is too large. However, if the calculation accuracy of the structure local is to be improved, it is necessary to subdivide local cells, which are referred to as target cells in this patent, and the remaining cells are peripheral cells, i.e., target cells representing cells that need to be subdivided a plurality of times. The density of the local unit of the unit is increased after subdivision, and the calculation accuracy of the fatigue crack propagation stress intensity factor can be improved when the method is applied to the division of the unit at the front edge of the crack.

A topology surface generally belongs to one cell or two cells, and the topology surface has three states according to the cell type to which the topology surface belongs: the topological surfaces in the three states can be divided into three types: a target cell face F2, a peripheral cell face F0, and a transition cell face F1.

Fig. 6 is a schematic diagram of the classification of the topology unit according to the embodiment of the present invention. A topological unit comprises 6 topological surfaces, and the topological unit can be classified according to the type of the topological surfaces, wherein the classification condition is as follows: all-encompassing target cell plane S0, mixed state S1, mixed state S2 (fig. 6, S2), and all-encompassing peripheral cell plane S3. Different types of topological elements can be classified using different templates. Wherein all faces of S0 are in F0 state; one surface of S1 is in F2 state, and the rest surfaces are in F0 state; 2 adjacent surfaces in the S2 are in F2 states, and the rest surfaces are in F0 states; all faces in S3 were in the F2 state.

Fig. 7 is a schematic diagram of face and edge subdivision provided by an embodiment of the present invention. The topological edge subdivision process can be divided into two states of E1 and E2, the state E2 is subdivided into 1-3 states (1-3 states are subdivided into a subdivision template which is conventionally used in the field, namely, the linear topological edge is subdivided into 3 sections in the invention), and the state E1 does not need to be subdivided; the topological surface can be divided into five states of G1, G2, G3, G4 and G5, wherein the G1 state comprises 4 topological edges of E1 states; the G2 states include a topology ratio edge of 3E 1 states and 1E 2 states; the G3 states include a topology ratio edge of 2E 1 states and 2E 2 states; the G4 states include 1 topological edge of the E1 state and 3 topological ratio edges of the E2 state; the G5 state includes topological specific edges of 4E 2 states, where the G3 and G4 states are not present. And the topological surfaces in different states are subdivided according to different subdivision strategies.

Fig. 8 is a schematic view of a hexahedral cell transition template according to an embodiment of the present invention. And classifying according to 6 topological surface states contained in the topological units, wherein the topological units of different types of topological units adopt different subdivision methods. The topology units are divided into S0, S1, S2 and S3. And the topological units in different states are subdivided by using different subdivision strategies, and the topological units are driven to be subdivided through a topological surface.

Fig. 9 and 10 are schematic diagrams of multi-level subdivision provided by the embodiment of the present invention. The unit topology data updating is used for multilevel subdivision, and the implementation steps of the multilevel subdivision are as follows:

(1) and the new topological structure generated after the unit is stored in a new hierarchical database.

The edge and cell data of the cell in the initial case are stored in an initial hierarchical database; the edges, the surfaces and the units in the hierarchical database are subdivided according to the strategy to generate new topological units, topological edges and topological surfaces, and the newly generated unit data need to be stored locally, so that a hierarchical database for storing new unit data is newly generated every time of subdivision. The hierarchical database is a hierarchical database in the prior art, and each subdivision is sequentially generated by new units, similar to the hierarchy.

(2) And newly generated nodes are uniformly stored in a node database, and the node topology structure database is globally shared. The subdivision process of the node database in the invention is a sharing function.

(3) The peripheral unit and the target unit are automatically identified.

In the invention, an initial unit (a topological unit before subdivision operation) is correspondingly marked, namely marked as a target unit or a peripheral unit, and the subdivided unit is a template unit; the subdivision is to subdivide the target unit, and the subdivided new unit inherits the mark of the target unit, namely the subdivided units are all the target units.

The recognition algorithm is executed here, described as:

(4) Subdividing again on the newly generated hierarchical database. The concrete contents of the subdivision are as follows:

one cell size is specified, each subdivided cell lateral size in the newly generated hierarchical database is compared with the cell size, and if the subdivided cell lateral size is larger than the specified cell size, the newly generated hierarchical database needs to be subdivided according to the subdivision operation.

(5) Turning to (1), multi-stage subdivision is possible. The multilevel subdivision has the same meaning as the subdivision in the step (4), and the subdivision can be carried out again according to the actual encryption requirement. Thereby increasing the cell density and achieving more efficient encryption processing.

While the invention has been described with reference to specific embodiments, the invention is not limited thereto, and those skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A full hexahedron unit encryption method based on a finite element model is characterized by comprising the following steps:

1) carrying out topological data reconstruction on the hexahedral unit based on the finite element model;

2) constructing topological unit structure data of the converted units;

3) dividing the topological unit structure data into a target unit and peripheral units, marking each unit, wherein the target unit is a unit needing to be subdivided, expressing each unit by adopting unit topological structure data, and acquiring a transition unit of a hexahedral unit;

4) and updating the topological data of the unit model, repeatedly performing multi-stage subdivision, increasing the unit density and finishing encryption processing.

2. A method for encrypting a full hexahedral element based on a finite element model according to claim 1, wherein the step 1) comprises the following steps:

11) defining a hexahedral cell as quad8, the topology of the cell consisting of six topological faces;

12) defining each topology surface as a face, wherein each face consists of four topology edges;

13) defining each topological edge as an edge, wherein each edge consists of two topological points;

14) defining each topological point as a vertex, wherein the topological point is a minimum topological element, represents a space node position and has a unique id;

15) all topological elements are shared, so that two adjacent hexahedral units share one topological surface.

3. The finite element model-based full hexahedral element encryption method according to claim 1, wherein in the step 2), the creation process of the element topology structure data comprises the following steps:

21) opening a unit description file and reading nodes, if the reading of the node data is completed, reading the unit data and executing the next step, otherwise, reading the node id, reading a coordinate system where the node is located and creating a topology node vertex;

22) judging whether the reading of the unit data is finished, if so, finishing all the steps, otherwise, reading the unit id and reading the eight node ids;

23) creating a topological surface according to the eight nodes, and decomposing the eight nodes into six groups of topological surface node groups, wherein each node group comprises four nodes;

24) creating six groups of topological surfaces according to a topological surface creating method;

24) and finishing the creation of a topological unit according to the steps.

4. A method for encrypting a full hexahedral element based on a finite element model according to claim 3, wherein in the step 24), the method for creating the topological surface specifically comprises the following steps:

241) searching a topology node vertex according to the input four topology node ids to obtain four topology nodes;

242) grouping the four topological nodes to obtain four topological edge input data, wherein each topological edge input data consists of two topological nodes;

243) and performing topology edge creation on each topology edge input data, and completing the creation of the topology edge.

5. The finite element model-based full hexahedron unit encryption method according to claim 4, wherein in the step 243), the specific process of creating the topological edge includes:

a) searching a corresponding topological edge according to input topological node data;

b) traversing the data structure of the topological edges, and searching whether the topological edges meet the equivalence of the topological nodes and the input nodes;

c) if the topological edge described in the step b) exists, directly returning the shared topological edge of the topological edge;

d) if the topological edge described in the step b) does not exist, a new topological edge is created according to the two input nodes;

e) and completing the creation of the topological edge.

6. A method for encrypting a hexahedral cell based on a finite element model according to claim 2, wherein in the step 3), each of the topology surfaces is divided into a target cell surface, a peripheral cell surface and a transition cell surface from the viewpoint of the topology surface, and the topology surfaces are divided into a full inclusion target cell surface, a first mixed state surface, a second mixed state surface and a full inclusion peripheral cell surface from the viewpoint of the state of the target surface included in the topology cell.

7. A finite element model-based all-hexahedral element encryption method according to claim 6, wherein all the faces all including the target element face are peripheral element faces, one of the first mixed-state faces is a target element face and the remaining faces are peripheral element faces, two adjacent faces of the second mixed-state face are target element faces and the remaining faces are peripheral element faces, and all the faces all including the peripheral element faces are target element faces.

8. The finite element model-based full hexahedron unit encryption method according to claim 2, wherein in step 3), the specific contents of the transition unit of the obtained hexahedron unit are: and sequentially carrying out topology edge subdivision, topology surface subdivision and topology unit subdivision on the topology elements.

9. The finite element model-based full hexahedron unit encryption method of claim 8, wherein the specific content of the topological edge subdivision is as follows: subdividing each topological edge into an E1 state or an E2 state, wherein the E1 state does not contain a target unit surface, and subdividing the state by 1-3, the E2 state contains the target unit surface and is a non-subdivided state; the specific content of the topology surface subdivision is as follows: the topological surfaces are divided into one of the following three states of G1, G2 and G5, wherein the G1 state comprises four topological edges of the E1 state, the G2 state comprises three topological edges of the E1 state and one topological edge of the E2 state, and the G5 state comprises four topological edges of the E2 state.

10. A method for encrypting a full hexahedral element based on a finite element model according to claim 1, wherein the step 4) comprises the following steps:

41) storing the new topological data generated in the steps 1) to 3) in a new hierarchical database, wherein the new topological data comprises newly generated topology edges, topology surfaces and topological units;

42) uniformly storing the newly generated nodes in a node topology database, and enabling the node topology database to carry out global sharing;

43) automatically identifying the peripheral unit and the target unit;

44) subdividing the newly generated node topology database again;

45) turning to step 41), multi-stage subdivision is carried out, the cell density is increased again, and encryption processing is achieved.