Aero-engine disk triangular-quadrilateral two-dimensional hybrid grid stress partition method

Abstract

The present application relates to the technical field of aero-engine probability damage tolerance evaluation, and particularly relates to a triangular quadrilateral two-dimensional hybrid grid stress partitioning method for aero-engine wheel disc, comprising the following steps: obtaining finite element stress analysis result data of the wheel disc component, preprocessing, and obtaining preprocessed input data; obtaining a convex hull boundary formed by all nodes by using the Graham method, and obtaining a graph boundary based on the convex hull boundary; obtaining stress pre-grouping and element type pre-grouping of each grid element; establishing an element adjacency graph based on the adjacent relationship of each grid element; searching the element adjacency graph by using a breadth-first search algorithm, and dividing grid elements with the same stress pre-grouping and the same element type pre-grouping into the same partition; and finally obtaining partition numbers of each grid element as the wheel disc stress partitioning result; the present application can improve the calculation accuracy and efficiency of the wheel disc stress partitioning.

Description

Technical Field

[0001] This invention relates to the field of probabilistic damage tolerance assessment technology for aero-engines, specifically to a two-dimensional hybrid mesh stress zoning method using a triangular quadrilateral grid for aero-engine rotor disks. Background Technology

[0002] Life-limited components (LLCs) in aero-engines refer to rotors and major stator structural components whose failure could lead to catastrophic engine consequences. For example, turbine disk failure could generate non-contained high-energy debris. Even a very low probability defect in an LCC can severely threaten the safety of the engine and, consequently, the entire aircraft. Therefore, the aerospace industry has proposed an additional life management process to ensure the safety of LCCs. The core of this process is to consider the extremely low probability of defects occurring in LCCs and conduct a probabilistic failure risk assessment. This assessment method combines factors such as the original defect distribution, disk load, and boundary conditions to calculate the failure risk of aero-engine LCCs within a certain flight cycle, thereby evaluating the safety of the LCC design scheme.

[0003] Taking a roulette wheel as an example, the common probabilistic failure risk assessment process both domestically and internationally is as follows: A finite element model of the three-dimensional roulette wheel component is established; based on the stress, temperature, and distance to the surface of the finite element, the component is divided into partitions; the failure probability of each partition is calculated separately; and finally, the overall failure risk of the roulette wheel is calculated. Due to the large number of elements, current mainstream failure probability calculation methods (such as the Monte Carlo method) are inefficient. Therefore, calculating the failure risk of each element sequentially would be too time-consuming. To balance calculation accuracy and time cost, a strategy is adopted to group elements with similar failure risks together and calculate them in a single step.

[0004] The latest partitioning algorithms, both domestically and internationally, are stress partitioning algorithms for two-dimensional finite element quadrilateral meshes. They determine the node type by the number of connected elements, thereby identifying the element type. Pre-partitioning is performed based on stress, temperature, and element type. Finally, geometric continuity is used to refine the partitions, ensuring geometric connectivity of elements within each partition. This provides efficient and accurate partitioning results for probabilistic failure risk assessment of roulette wheels. Summary of the Invention

[0005] In view of the above problems, the present invention provides a two-dimensional hybrid mesh stress partitioning method for aero-engine wheel disks using triangular quadrilaterals, which solves the technical problem that the calculation accuracy and efficiency of wheel disk stress partitioning in the prior art are not high enough.

[0006] This invention provides a two-dimensional hybrid mesh stress partitioning method using triangular and quadrilateral grids for aero-engine rotor disks, comprising the following steps: Step S1: Obtain the finite element stress analysis results data of the wheel component, including the coordinates of each node of the principal stress orthogonal plane of the wheel component, the maximum principal stress value and the corresponding mesh element number; wherein each node of the principal stress orthogonal plane forms a triangular or quadrilateral mesh element; The finite element stress analysis results data are preprocessed to obtain preprocessed input data; Step S2: Process the preprocessed input data using the Graham method to obtain the convex hull boundary formed by all nodes, and perform a search based on the convex hull boundary to obtain the graphic boundary; Step S3: Obtain stress pre-grouping for each mesh element based on the preprocessed input data; obtain element type pre-grouping for each mesh element based on the graphical boundary; establish an element adjacency graph based on the adjacency relationship of each mesh element. Step S4: Use the breadth-first search algorithm to search the cell adjacency graph and divide the mesh cells with the same stress pregrouping and the same cell type pregrouping into the same partition; Finally, the partition number of each grid cell is obtained, which serves as the disk stress partitioning result.

[0007] Preferably, in step S1, the step of preprocessing the finite element stress analysis result data to obtain preprocessed input data specifically includes: (1) Perform data cleaning to remove duplicate node data rows and Gaussian integration points of mesh elements in the finite element stress analysis results data; (2) Using the coordinate system transformation method, the three-dimensional coordinates of each node are transformed to the principal stress orthogonal plane to obtain the planar coordinates; (3) Sort the vertices of the triangle and convex quadrilateral units clockwise or counterclockwise; for the triangle, the three vertices are directly sorted clockwise or counterclockwise; for the quadrilateral, the four vertices are sorted clockwise or counterclockwise by the method of determining the adjacency of two vertices. (4) Obtain the preprocessed input data, including: node number, coordinates, grid cell number corresponding to the node, and the number of the vertex nodes of the grid cell corresponding to the node arranged clockwise or counterclockwise.

[0008] Preferably, step S2 specifically includes: Step S2-1: Process all nodes in the preprocessed input data using the Graham method to obtain the node sequence of the convex hull boundary. ; Step S2-2: Establish a closed graph G based on the node information in the preprocessed input data. The nodes of the closed graph G are the nodes in the preprocessed input data, and the edges of the closed graph G are the edges between the nodes in the preprocessed input data. Step S2-3: Based on the node sequence of the convex hull boundary and the closed graph G, perform a recursive search to obtain the graphic boundary that represents the graphic formed by all nodes.

[0009] Preferably, step S2-3 specifically includes: Step S2-3-1: Node sequence based on convex hull boundary Obtain the initial points and initial boundaries of the closed graph G, specifically including: The first node in the convex hull boundary node sequence is The second node is ,Will Set the initial point as described above, and obtain the closed graph G containing the points that are connected to the initial point. The set of all connected nodes, from... Find the node in the set of all connected nodes. Requires nodes Satisfaction makes Minimum, will Set as the initial boundary; Step S2-3-2: Based on the initial boundary, perform multiple recursive iterations to finally obtain the graphic boundary, specifically including: Establish the boundary recursion process, including: using Derivation The boundary recursion method, where and These represent the nodes on the boundary of the graphic, whose numbering sequences are j-1, j, and j+1 respectively, arranged counterclockwise. The graphic boundary node sequence is initialized with the first two nodes in the convex hull boundary node sequence. Then, the boundary recursion process is applied sequentially to two consecutive nodes in the graphic boundary node sequence until the complete graphic boundary node sequence is obtained. ; sequence of graphical boundary nodes The nodes in the graph are arranged counterclockwise to form the boundary of the graphic.

[0010] Preferably, in step S2-3-2, the use of Derivation The boundary recursion methods specifically include: Based on the acquisition of nodes in a closed graph G Connected by edges The set of nodes is ,in, They represent the 1st, 2nd, and 3rd respectively. Individual and Node Connected nodes; Enumerate all non-zero values ​​in the set Let the currently enumerated node be . , Indicates the k-th node Connected nodes; set up , They represent , The unit direction vector, , Representing nodes respectively To the node ,node To the node The vector, , They represent , The coordinates; Obtained through geometric transformation arrive The counterclockwise angle ; At each of the enumerated nodes corresponding Among the values, select the one that makes The smallest node is used as the recursive result.

[0011] Preferably, step S3 specifically includes: Step S3-1: Determine the stress pre-grouping of the mesh elements based on the stress values ​​at all vertices and the stress interval values ​​of the mesh elements; Step S3-2: Determine the cell type pregrouping of the mesh cells based on whether the vertices of the mesh cells belong to the nodes in the graphic boundary; Step S3-3: Use each grid cell in the preprocessed input data as a node set of the cell adjacency graph. If two units share a common edge, then a connection is established. The undirected edges corresponding to the nodes in the equation are used to obtain the edge set. ,Depend on and Establish a cell adjacency graph to describe the geometric continuity between cells. ; Each element has a record of the element stress pregrouping and element type pregrouping.

[0012] Preferably, in step S3-1, the step of determining the stress pregrouping of the mesh elements includes: Mesh elements that meet the following requirements are assigned to multiple stress pregroups: The stresses of mesh elements within the same stress pregroup all belong to a width of A defined and continuous stress value; and each mesh element belongs to one and only one uniquely defined stress pregroup; Step S3-2, the step of determining the element type pre-grouping of the mesh elements, includes: Mesh cells that meet the following requirements will be assigned to multiple pregroups of cell types: If a unit has at least one node appearing in the sequence of graphical boundary nodes... If a vertex in the array is selected, its element type is pregrouped as a surface element; otherwise, its element type is pregrouped as an internal element.

[0013] Preferably, step S4 specifically includes: Step S4-1: Set the variable num to represent the partition number currently being maintained; complete the initialization, including creating an empty queue. Assign num the value of 0, and set the graph... All nodes in the list are marked as unvisited; Step S4-2, for the graph The set of nodes in If a node in the queue has not yet been visited, then the current node is added to the queue. It also marks the current node as visited and updates the partition number num to num+1; Step S4-3: Retrieve the current queue The first element of the team , the first element of the team Set the partition number of the corresponding unit to num, and iterate through all units that are related to it. In the figure A node with a connected edge, if it is connected to... In the figure The first edge in the middle l Nodes When the following conditions are met: Not yet visited. Corresponding stress pregrouping and element type pregrouping The corresponding grouping information is exactly the same; Then Add to the queue and Marked as already visited; Step S4-4: Return to step S4-2 until the graph has been traversed. The set of nodes in All nodes in the middle.

[0014] Preferably, a stress partitioning diagram is drawn based on the partitioning number of each grid cell and displayed on the user interface.

[0015] Compared with the prior art, the present invention has at least the following beneficial effects: (1) By systematically preprocessing the finite element stress analysis results, this invention can effectively support the stress partitioning requirements of a mixture of triangular and convex quadrilateral element shapes, breaking the limitation of traditional methods on single element shapes, improving the applicability and flexibility of stress partitioning, and meeting the engineering needs of more complex mesh structures.

[0016] (2) The present invention uses the Graham algorithm to automatically obtain the convex hull boundary formed by all nodes, and combines it with the graphical boundary search based on the convex hull boundary to realize autonomous boundary recognition without additional input of mesh topology boundary information, which simplifies the data preparation process and improves computational efficiency and operational convenience.

[0017] (3) The present invention establishes a cell adjacency graph and uses a breadth-first search algorithm to perform joint partitioning of the mesh cells based on stress and cell type, ensuring that the partitioning results are highly accurate and consistent with the calculation shape of the FAA airworthiness advisory circular. This verifies that the method has high computational efficiency while maintaining partitioning accuracy, and its method framework has the potential to be extended to the partitioning of three-dimensional models, and has good scalability. Attached Figure Description

[0018] The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of the invention.

[0019] Figure 1 The flowchart shows the stress partitioning method of the triangular quadrilateral two-dimensional hybrid mesh for aero-engine wheel disk provided by the present invention. Detailed Implementation

[0020] To better understand the above-described objectives, features, and advantages of the present invention, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments of the present invention and the features thereof can be combined with each other. Furthermore, the present invention can be implemented in other ways different from those described herein; therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0021] This invention discloses a computational geometry-based stress partitioning algorithm for hybrid triangular and convex quadrilateral meshes. First, the mesh and stress data obtained from finite element stress analysis are preprocessed to obtain the correspondence between elements and nodes, enabling stable extraction of corresponding nodes for triangular and convex quadrilateral elements in clockwise or counterclockwise order. The Graham algorithm for computational geometry is used to solve for the convex shell boundaries of the nodes, thereby recursively obtaining the mesh boundaries. Elements are pre-partitioned based on stress similarity and near-surface partitioning principles. An undirected graph is established based on element adjacency relationships, and a breadth-first search algorithm is used to maintain the geometric continuity of the partitions, resulting in a final partitioning result that satisfies the three principles of stress similarity, near-surface partitioning, and geometric continuity. The time complexity is O(log n). .

[0022] Compared to existing partitioning methods, the method provided by this invention can meet the stress partitioning requirements of mixed triangular and convex quadrilateral element shapes, requires no additional input mesh topological boundary information, and has higher computational efficiency. Furthermore, the partitioning results obtained are consistent with the shapes of the examples provided in the FAA airworthiness advisory circular. In addition, this invention has the capability to be extended to partitioning three-dimensional models.

[0023] To illustrate the effectiveness of the method proposed in this invention, the following detailed description of the above technical solution is provided through a specific embodiment, such as... Figure 1 As shown, a two-dimensional hybrid mesh stress zoning method using triangular and quadrilateral grids for aero-engine rotor disks is disclosed. The specific implementation steps are as follows: Step S1: Obtain the finite element stress analysis results data of the wheel component, including the coordinates of each node of the principal stress orthogonal plane of the wheel component, the maximum principal stress value and the corresponding mesh element number; wherein each node of the principal stress orthogonal plane constitutes a triangular or quadrilateral mesh element; preprocess the finite element stress analysis results data to obtain the preprocessed input data; In this step, the present invention first obtains the finite element stress analysis results data of the wheel component. In some embodiments, the wheel can be modeled and solved using finite element methods, and the resulting principal stress orthogonal plane mesh data can be used as the finite element stress analysis results data. The data includes various information of all nodes on the principal stress orthogonal plane, wherein each node on the principal stress orthogonal plane constitutes a triangular or quadrilateral mesh element. The form of the node data is shown in Table 1.

[0024] Table 1

[0025] in, Indicates the node number. Represents a node The three-dimensional coordinates Indicates the cell number to which the node belongs. Represents a node The maximum principal stress value at that location.

[0026] In this step, the present invention preprocesses the finite element stress analysis results data, including the following steps: (1) Perform data cleaning to remove duplicate node data rows and Gaussian integration points of mesh elements in the finite element stress analysis results data to ensure the uniqueness of node data.

[0027] (2) Since all nodes are located in the principal stress orthogonal plane, a coordinate transformation method is used to transform the three-dimensional coordinates to the principal stress orthogonal plane. For example, the coordinates of all nodes can be converted to Same coordinates, then obtain , coordinate.

[0028] (3) Sort the vertices of the triangle and convex quadrilateral units clockwise or counterclockwise. The three vertices of a triangle, arranged arbitrarily, will necessarily satisfy either a clockwise or counterclockwise order. For a quadrilateral, this invention first obtains the order of its four vertices to ensure that they are arranged clockwise or counterclockwise in the processed data. The method for obtaining the order of the four vertices described in this invention can employ a two-vertex adjacency determination method or other determination methods; this invention does not limit the specific method for obtaining the order of the four vertices.

[0029] The following example illustrates a method for determining the adjacency of two vertices: Suppose the four vertices of a convex quadrilateral are... as well as Their adjacency relationship and order in the quadrilateral are unknown. Taking points A and B as an example, if the two points are adjacent, then points C and D are both located on the same side of the line segment AB. Let the function be: (1) in, This represents the decision function. Let A and B represent the coordinates of the two vertices, respectively. Represents the coordinates of the vertex.

[0030] (2) in, Let A and B represent the coordinates of vertices C and D, respectively. If the above inequality holds, then vertices A and B are considered adjacent vertices.

[0031] The following example illustrates the steps of obtaining the order of the four vertices using the two-vertex adjacency determination method: For a convex quadrilateral with vertices ABCD, the two-vertex adjacency determination method first determines whether AB is adjacent, then determines whether AC is adjacent for both cases where AB is adjacent and not adjacent, ultimately obtaining a clockwise or counterclockwise order. Specifically: if AB is adjacent and AC is adjacent, the order of the four vertices is ABDC; if AB is adjacent and AC is not adjacent, the order of the four vertices is ABCD; if AB is not adjacent and AC is adjacent, the order of the four vertices is ACBD. Since ABCD is the four vertices of a quadrilateral, the above three cases cover all possible order determinations.

[0032] Table 2

[0033] Based on the method of determining the adjacency of two vertices, the process for determining the order of the four vertices of a convex quadrilateral, as shown in Table 2, can be obtained.

[0034] After the above preprocessing is completed, the data shown in Table 3 is obtained as the preprocessed input data. The three columns are node information, including the node number and coordinate information; the five columns are cell information, including the cell number and the cell vertex numbers arranged clockwise or counterclockwise.

[0035] Table 3

[0036] in, , , , These represent the vertex numbers of the four units arranged clockwise or counterclockwise, respectively.

[0037] Through this step, the present invention obtains structured node and corresponding unit data as preprocessed input data for further analysis and processing.

[0038] Step S2: Process the preprocessed input data using the Graham method to obtain the convex hull boundary formed by all nodes, and perform a search based on the convex hull boundary to obtain the graphic boundary; In this step, the present invention uses the Graham method of computational geometry to obtain the convex shell boundary formed by all nodes on the principal stress orthogonal plane, and then uses the convex shell boundary to recursively search for the graphic boundary representing the graphic formed by all nodes.

[0039] The steps for obtaining the convex hull boundary using the Graham method are described in detail below.

[0040] (1) For the set of all nodes, find the node with the smallest x and y coordinates as the starting point, prioritizing the node with the smallest y value. If there are multiple nodes with the smallest y value, select the node with the smallest x value as the starting point. The coordinates of the starting point are: .

[0041] (2) Enumerate all nodes except the starting point, and the coordinates of the current node are: Then the deflection angle of the current node will be... Defined as the angle between the vector pointing from the starting point to the current node and the direction vector of the positive x-axis. The expression is: (3) in, This indicates the deflection angle of the current node. Represents the coordinates of the current node.

[0042] Apply deflection angle to all nodes Sort the nodes from smallest to largest to obtain the node number sequence. .

[0043] (3) Create a new stack data structure, and set the starting point and... The first node in the stack is pushed onto the stack; (4) From Starting from the second node, we enumerate all the nodes in it one by one. For the current node, set up This represents the coordinates of the node preceding the top node of the stack. Indicates the coordinates of the top node of the stack. This represents the coordinates of the currently enumerated node. If the vector... With vector If the product vector is pointing downwards (negative), the top element of the stack is popped; otherwise, the top element is retained. This invention does not impose specific restrictions on the determination method.

[0044] One possible decision expression is: (4) Then, push the currently enumerated node onto the stack until there are fewer than two elements in the stack or the current top element does not need to be popped. Finally, arrange the remaining nodes in the stack in order to obtain the node sequence of the convex hull boundary. .

[0045] Subsequently, the present invention processes the node information in the preprocessed input data to establish a closed graph G. The nodes of the closed graph G are the nodes in the preprocessed input data, and the edges of the closed graph G are the edges between the nodes in the preprocessed input data.

[0046] In this step, the present invention performs a recursive search based on the node sequence of the convex hull boundary and the closed graph G to obtain the graphic boundary representing the graphic formed by all nodes, as described in detail below.

[0047] (1) Locate the initial point and initial boundary of the closed graph G based on the node sequence of the convex hull boundary. The first node in the convex hull boundary node sequence is The second node is , and There is a connecting edge between them on the boundary of the convex shell.

[0048] In this step, Set the initial point as stated, and then obtain the points in the closed graph G that are connected to the initial point. The set of all connected nodes, from... Find the node in the set of all connected nodes. Requires nodes Satisfaction makes Minimum, will Set as the initial boundary.

[0049] (2) Derive the next edge based on the current boundary, and search sequentially to obtain the boundary of the figure.

[0050] This step sets up the usage Derivation The boundary recursion method, where, and Let j-1, j, and j+1 be the nodes on the boundary of the figure, arranged counterclockwise. The recursive method is as follows: Based on the acquisition of nodes in a closed graph G Connected by edges The set of nodes is ,in, They represent the 1st, 2nd, and 3rd respectively. Individual and Node Connected nodes. Enumerate all nodes in the set that are not... Let the currently enumerated node be . , Indicates the k-th node Connected nodes.

[0051] Vectors can be calculated using the following expression. , Unit direction vector: (5) (6) set up , They represent , The unit direction vector, , Representing nodes respectively To the node ,node To the node The vector, , They represent , The coordinates.

[0052] Solve the following system of equations: (7) Obtained through geometric transformation arrive The counterclockwise angle The expression is: (8) At each of the enumerated nodes corresponding Among the values, select the one that makes The smallest node is used as the node to be recursively processed. .

[0053] The above steps describe the use of Derivation The boundary recursion method; the process of traversing and finding the boundary of the figure is as follows: First, Assigned value Then Assigned value Repeat the boundary recursion method described above, utilizing... Calculated ,use Calculated ...and so on, until a complete sequence of graphical boundary nodes is obtained. .

[0054] Finally, the sequence of graph boundary nodes The nodes in the graph are arranged counterclockwise to form the boundary of the graphic.

[0055] Step S3: Obtain stress pre-grouping for each mesh element based on the preprocessed input data; obtain element type pre-grouping for each mesh element based on the graphical boundary; establish an element adjacency graph based on the adjacency relationship of each mesh element; In this step, the present invention obtains corresponding stress pregrouping and element type pregrouping for each mesh element in the preprocessed input data.

[0056] (1) Stress pregrouping Set stress range This refers to the width of each stress pregroup. Dividing all mesh elements within the principal stress orthogonal plane into several stress pregroups requires the following: 1. The stresses of mesh elements within the same stress pregroup all belong to a width of... 1. A defined and continuous stress value; 2. Each mesh element belongs to one and only one uniquely defined stress pre-partition.

[0057] The stress of each mesh element is set to the maximum value of the stresses at all its vertices. For example, the element numbered... Stress of the mesh element Represents grid cells The maximum value of all vertex stresses. The maximum value of all vertex stresses within the entire principal stress orthogonal plane is... This invention does not restrict the specific numbering method of pregrouping, as long as the above requirements are met. An expression for calculating the stress pregrouping number of a mesh element is as follows: (9) in, Indicates the stress pregrouping number of the mesh element. This represents the stress value at the point where the stress reaches its maximum value within the plane of orthogonal principal stresses. Represents grid cells The maximum value of all vertex stresses, Indicates the stress range.

[0058] (2) Pregrouping of unit types Mesh elements are pre-grouped into two types based on the type of their four vertices: internal elements and surface elements. The specific classification rules are as follows: 1. If an element has at least one vertex that appears in the sequence of boundary nodes of the graph... 1. If the vertex is in the range, its element type is pregrouped as surface element; 2. The element type of other elements is pregrouped as internal element.

[0059] This invention does not restrict the specific method of pre-grouping and numbering the cell type, as long as the above rules are met. A calculation expression for the pre-grouping and numbering of a grid cell type is as follows: (10) in, The cell type pregrouping number indicates the type of the grid cell. Representing grid cells respectively The 1st, 2nd, 1 node Represents grid cells The total number of nodes, Represents the sequence of boundary nodes of a graph. This indicates taking the intersection. This represents the empty set.

[0060] (3) Establish a cell adjacency graph based on the adjacency relationship of each grid cell. In this step, each grid cell in the preprocessed input data is used as the node set of the cell adjacency graph. If two units share a common edge, then a connection is established. The undirected edges corresponding to the nodes in the equation are used to obtain the edge set. ,Depend on and Establish a cell adjacency graph to describe the geometric continuity between cells. Cell adjacency graph Each element has a record of the element stress pregrouping and element type pregrouping.

[0061] Step S4: Use the breadth-first search algorithm to search the cell adjacency graph and divide the mesh cells with the same stress pregrouping and the same cell type pregrouping into the same partition; Finally, the partition number of each grid cell is obtained, which serves as the disk stress partitioning result.

[0062] In this step, the graph created in the previous step is... Perform the following operations to maintain the final partition information.

[0063] (1) Set the variable num to represent the partition number currently being maintained; complete the initialization, including creating an empty queue. Assign num the value of 0, and set the graph... All nodes in the list are marked as unvisited.

[0064] (2) For the graph The set of nodes in If a node in the queue has not yet been visited, then the current node is added to the queue. It also marks the current node as visited and updates the partition number num to num+1; (3) Retrieve the current queue The first element of the team , the first element of the team Set the partition number of the corresponding unit to num, and iterate through all units corresponding to it. In the figure A node with a connected edge, if it is connected to... In the figure The first edge in the middle l Nodes When the following conditions are met: Not yet visited. The corresponding element stress pregrouping and element type pregrouping are... They are exactly the same; Then Add to the queue and Marked as already visited; Return to step (2) until the graph has been traversed. The set of nodes in All nodes in the middle.

[0065] After completing the above steps, the partition information of the maintained unit is output, and the partitioning result based on stress similarity, near-surface partitioning and geometric continuity principles obtained in this invention can be obtained.

[0066] In some embodiments, the graphic boundaries obtained in the preceding steps, together with the partitioning results obtained in this step, can be used to generate four visualization results: a unit schematic diagram, a surface unit schematic diagram, a stress cloud diagram, and a stress partitioning schematic diagram, which are then displayed on the user interface.

[0067] In some embodiments, a schematic diagram of the elements can be drawn from the node and element information in the preprocessed input data to represent the finite element mesh structure; a schematic diagram of the surface elements can be drawn from the graphical boundaries; a stress cloud map can be drawn from the stress information of the nodes and elements; and a schematic diagram of the stress partition can be drawn from the partition information to which the elements belong.

[0068] While the specific embodiments of the present invention depict actions or steps in a particular order, this should be understood as requiring such actions or steps to be performed in the shown specific order or sequential order, or requiring all illustrated actions or steps to be performed to achieve the desired result. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single implementation. Conversely, various features described in the context of a single implementation may also be implemented individually or in any suitable sub-combination in multiple implementations. The above descriptions are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention.

[0069] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for stress partitioning of a two-dimensional hybrid mesh consisting of a triangular and quadrilateral grid on an aero-engine wheel disk, characterized in that, Includes the following steps: Step S1: Obtain the finite element stress analysis results data of the wheel component, including the coordinates of each node of the principal stress orthogonal plane of the wheel component, the maximum principal stress value and the corresponding mesh element number; wherein each node of the principal stress orthogonal plane forms a triangular or quadrilateral mesh element; The finite element stress analysis results data are preprocessed to obtain preprocessed input data; Step S2: Process the preprocessed input data using the Graham method to obtain the convex hull boundary formed by all nodes, and perform a search based on the convex hull boundary to obtain the graphic boundary; Step S3: Obtain the stress pregrouping of each mesh element based on the preprocessed input data; Based on the graphical boundaries, the cell types of each grid cell are pre-grouped. Establish a cell adjacency graph based on the adjacency relationships of each grid cell; Step S4: Use the breadth-first search algorithm to search the cell adjacency graph and divide the mesh cells with the same stress pregrouping and the same cell type pregrouping into the same partition; Finally, the partition number of each grid cell is obtained, which serves as the disk stress partitioning result.

2. The two-dimensional hybrid mesh stress zoning method for aero-engine wheel disks using triangular and quadrilateral grids according to claim 1, characterized in that, In step S1, the step of preprocessing the finite element stress analysis results data to obtain preprocessed input data specifically includes: (1) Perform data cleaning to remove duplicate node data rows and Gaussian integration points of mesh elements in the finite element stress analysis results data; (2) Using the coordinate system transformation method, the three-dimensional coordinates of each node are transformed to the principal stress orthogonal plane to obtain the planar coordinates; (3) Sort the vertices of the triangle and convex quadrilateral units clockwise or counterclockwise; for the triangle, the three vertices are directly sorted clockwise or counterclockwise; for the quadrilateral, the four vertices are sorted clockwise or counterclockwise by the method of determining the adjacency of two vertices. (4) Obtain the preprocessed input data, including: node number, coordinates, grid cell number corresponding to the node, and the number of the vertex nodes of the grid cell corresponding to the node arranged clockwise or counterclockwise.

3. The two-dimensional hybrid mesh stress zoning method for aero-engine wheel disks using triangular and quadrilateral grids according to claim 2, characterized in that, Step S2 specifically includes: Step S2-1: Process all nodes in the preprocessed input data using the Graham method to obtain the node sequence of the convex hull boundary. ; Step S2-2: Establish a closed graph G based on the node information in the preprocessed input data. The nodes of the closed graph G are the nodes in the preprocessed input data, and the edges of the closed graph G are the edges between the nodes in the preprocessed input data. Step S2-3: Based on the node sequence of the convex hull boundary and the closed graph G, perform a recursive search to obtain the graphic boundary that represents the graphic formed by all nodes.

4. The two-dimensional hybrid mesh stress zoning method for aero-engine wheel disks using triangular and quadrilateral grids according to claim 3, characterized in that, Step S2-3 specifically includes: Step S2-3-1: Node sequence based on convex hull boundary Obtain the initial points and initial boundaries of the closed graph G, specifically including: The first node in the convex hull boundary node sequence is The second node is ,Will Set the initial point as described above, and obtain the closed graph G containing the points that are connected to the initial point. The set of all connected nodes, from... Find the node in the set of all connected nodes. Requires nodes Satisfaction makes Minimum, will Set as the initial boundary; Step S2-3-2: Based on the initial boundary, perform multiple recursive iterations to finally obtain the graphic boundary, specifically including: Establish the boundary recursion process, including: using Derivation The boundary recursion method, where and These represent the nodes on the boundary of the graphic, whose numbering sequences are j-1, j, and j+1 respectively, arranged counterclockwise. The graphic boundary node sequence is initialized with the first two nodes in the convex hull boundary node sequence. Then, the boundary recursion process is applied sequentially to two consecutive nodes in the graphic boundary node sequence until the complete graphic boundary node sequence is obtained. ; sequence of graphical boundary nodes The nodes in the graph are arranged counterclockwise to form the boundary of the graphic.

5. The two-dimensional hybrid mesh stress zoning method for aero-engine wheel disks using triangular and quadrilateral grids according to claim 4, characterized in that, In step S2-3-2, the use of Derivation The boundary recursion methods specifically include: Based on the acquisition of nodes in a closed graph G Connected by edges The set of nodes is ,in, They represent the 1st, 2nd, and 3rd respectively. Individual and Node Connected nodes; Enumerate all non-zero values ​​in the set Let the currently enumerated node be . , Indicates the k-th node Connected nodes; set up , They represent , The unit direction vector, , Representing nodes respectively To the node ,node To the node The vector, , They represent , The coordinates; Obtained through geometric transformation arrive The counterclockwise angle ; At each of the enumerated nodes corresponding Among the values, select the one that makes The smallest node is used as the recursive result. .

6. The two-dimensional hybrid mesh stress zoning method for aero-engine wheel disks using triangular and quadrilateral grids according to claim 5, characterized in that, Step S3 specifically includes: Step S3-1: Determine the stress pre-grouping of the mesh elements based on the stress values ​​at all vertices and the stress interval values ​​of the mesh elements; Step S3-2: Determine the cell type pregrouping of the mesh cells based on whether the vertices of the mesh cells belong to the nodes in the graphic boundary; Step S3-3: Use each grid cell in the preprocessed input data as a node set of the cell adjacency graph. If two units share a common edge, then a connection is established. The undirected edges corresponding to the nodes in the equation are used to obtain the edge set. ,Depend on and Establish a cell adjacency graph to describe the geometric continuity between cells. ; Each element has a record of the element stress pregrouping and element type pregrouping.

7. The two-dimensional hybrid mesh stress zoning method for aero-engine wheel disks using triangular and quadrilateral grids according to claim 6, characterized in that, Step S3-1, the step of determining the stress pregrouping of mesh elements includes: Mesh elements that meet the following requirements are assigned to multiple stress pregroups: The stresses of mesh elements within the same stress pregroup all belong to a width of A defined and continuous stress value; and each mesh element belongs to one and only one uniquely defined stress pregroup; Step S3-2, the step of determining the element type pre-grouping of the mesh elements, includes: Mesh cells that meet the following requirements will be assigned to multiple pregroups of cell types: If a unit has at least one node appearing in the sequence of graphical boundary nodes... If a vertex in the array is selected, its element type is pregrouped as a surface element; otherwise, its element type is pregrouped as an internal element.

8. The two-dimensional hybrid mesh stress zoning method for aero-engine wheel disks using triangular and quadrilateral grids according to claim 7, characterized in that, Step S4 specifically includes: Step S4-1: Set the variable num to represent the partition number currently being maintained; complete the initialization, including creating an empty queue. Assign num the value of 0, and set the graph... All nodes in the list are marked as unvisited; Step S4-2, for the graph The set of nodes in If a node in the queue has not yet been visited, then the current node is added to the queue. It also marks the current node as visited and updates the partition number num to num+1; Step S4-3: Retrieve the current queue The first element of the team , the first element of the team Set the partition number of the corresponding unit to num, and iterate through all units corresponding to it. In the figure A node with a connected edge, if it is connected to... In the figure The first connected edge l Nodes When the following conditions are met: Not yet visited. Corresponding stress pregrouping and element type pregrouping The corresponding grouping information is exactly the same; Then Add to the queue and Marked as already visited; Step S4-4: Return to step S4-2 until the graph has been traversed. The set of nodes in All nodes in the middle.

9. The two-dimensional hybrid mesh stress partitioning method for triangular quadrilateral grids of aero-engine wheel disks according to claim 8, characterized in that, Also includes: A stress partitioning diagram is drawn based on the partitioning number of each grid cell and displayed on the user interface.

Images