Fatigue hot spot prediction method and system

Abstract

The application relates to the fields of machinery and computers, and discloses a fatigue hot spot prediction method and system, which comprises the following steps: acquiring free surface grid data of a finite element model of a mechanical structure; determining fatigue results of each free surface according to the free surface grid data; searching for fatigue hot spots of the finite element model of the mechanical structure according to the fatigue results of each free surface; and outputting the fatigue hot spots of the finite element model of the mechanical structure. The application can reduce the test difficulty, improve the test precision, and reduce the test cost, thereby significantly improving the efficiency of the durability fatigue test.

Description

Technical Field

[0001] This application relates to the fields of mechanical and computer technology, and in particular to the technology of fatigue hot spot prediction. Background Technology

[0002] In physical prototype durability fatigue testing, the placement and orientation of strain sensors are crucial. There are two main types of strain sensors: unidirectional strain gauges and multidirectional strain rosettes. The cost of strain rosettes is much higher than that of strain gauges. Therefore, if the orientation of the strain gauges can be determined in advance, the testing cost can be greatly reduced.

[0003] Currently available approaches include, for example, estimating based on engineers' own engineering experience; or manually estimating the strain gauge placement position based on the maximum principal stress of the finite element results. However, this method is only applicable to simple cases with a single working condition and requires manually mapping the maximum principal stress onto the surface. In the real world, the stress conditions of products are complex and varied. Therefore, this method cannot accurately predict the fatigue hotspots of products under such complex loads. Another approach is to evaluate based on the maximum or minimum values ​​of the fatigue results. However, this method can only obtain one location. A model often has multiple hot zones, and each hot zone corresponds to a hot spot.

[0004] It is evident that current durability fatigue testing still faces several challenges, such as low testing accuracy, high testing costs, and high testing complexity. These factors significantly reduce the effectiveness of durability testing on physical prototypes. Summary of the Invention

[0005] The purpose of this application is to provide a fatigue hotspot prediction method and system that can reduce testing difficulty, improve testing accuracy, and reduce testing costs, thereby significantly improving the efficiency of durability fatigue testing.

[0006] This application discloses a method for predicting fatigue hotspots, including:

[0007] Obtain the free surface mesh data of the finite element model of the mechanical structure;

[0008] The fatigue result for each free surface is determined based on the free surface mesh data;

[0009] Based on the fatigue results of each free surface, the fatigue hotspots of the finite element model of the mechanical structure are searched.

[0010] Output the fatigue hotspots of the finite element model of the mechanical structure.

[0011] In a preferred embodiment, in the step of searching for fatigue hotspots in the finite element model of the mechanical structure based on the fatigue results of each free surface, if the fatigue hotspots are searched by comparing the fatigue result damage values, then the point with the largest damage value is the fatigue hotspot; if the fatigue hotspots are searched by comparing the fatigue result safety factor values, then the point with the smallest safety factor value is the fatigue hotspot.

[0012] In a preferred embodiment, in the step of searching for fatigue hotspots in the finite element model of the mechanical structure based on the fatigue results of each free surface, if the fatigue results of all adjacent free surfaces are greater than the fatigue result of the current free surface, then the current free surface is marked as Valley; if the fatigue results of all adjacent free surfaces are less than the fatigue result of the current free surface, then the current free surface is marked as Peak; if the fatigue results of adjacent free surfaces are both less than and greater than the fatigue result of the current free surface, then the current free surface is marked as Ignore; if the fatigue results of adjacent free surfaces are both less than and equal to the fatigue result of the current free surface, then the current free surface is marked as Pending; if the fatigue results of adjacent free surfaces are both greater than and equal to the fatigue result of the current free surface, then the current free surface is marked as Pending; if the current free surface is marked as Pending, a recursive method is used to continue checking neighboring nodes that are equal to it, until it is determined that the marking is not Pending, at which point the recursion ends.

[0013] In a preferred embodiment, the free surface refers to the element surface exposed on the surface of the finite element model.

[0014] In a preferred embodiment, the fatigue result of the free surface is the fatigue result of the center point of the free surface, or the fatigue result of the nodes of the free surface.

[0015] In a preferred embodiment, determining the fatigue result for each free surface based on the free surface mesh data includes the following sub-steps:

[0016] Based on the free surface mesh data, obtain the target element surface corresponding to each virtual patch in the finite element model of the mechanical structure;

[0017] Determine the position of the center of each target element surface in the first coordinate system and the stress / strain tensor matrix of the center of each target element surface relative to the first coordinate system;

[0018] A corresponding second coordinate system is established with the center of each target unit surface. The angle cosine matrix between the second coordinate system and each axis of the first coordinate system of the target unit surface is determined. Based on the angle cosine matrix between the second coordinate system and each axis of the first coordinate system of the target unit surface, and the stress / strain tensor matrix of the center of the target unit surface relative to the first coordinate system, the stress / strain tensor matrix of the center of the target unit surface relative to the second coordinate system is determined.

[0019] Based on the stress / strain tensor matrix of the center of each target unit surface in the corresponding second coordinate system, the fatigue durability life, and / or damage, and / or safety factor and the most damaged direction of the virtual patch corresponding to the target unit surface are determined as the fatigue result of the free surface corresponding to the target unit surface.

[0020] This application also discloses a fatigue hotspot prediction system, comprising:

[0021] The free surface network data acquisition module is used to acquire the free surface mesh data of the finite element model of the mechanical structure;

[0022] The fatigue result acquisition module is used to determine the fatigue result of each free surface based on the free surface mesh data;

[0023] The fatigue result search module is used to search for fatigue hotspots in the finite element model of the mechanical structure based on the fatigue results of each free surface.

[0024] The output module is used to output the fatigue hotspots of the finite element model of the mechanical structure.

[0025] In a preferred embodiment, when the fatigue result search module searches for fatigue hotspots in the finite element model of the mechanical structure based on the fatigue results of each free surface, if the fatigue hotspots are searched by comparing the fatigue result damage values, the point with the largest damage value is the fatigue hotspot; if the fatigue hotspots are searched by comparing the fatigue result safety factor values, the point with the smallest safety factor value is the fatigue hotspot.

[0026] In a preferred embodiment, when the fatigue result search module searches for fatigue hotspots in the finite element model of the mechanical structure based on the fatigue results of each free surface, if the fatigue results of all adjacent free surfaces are greater than the fatigue result of the current free surface, the current free surface is marked as Valley; if the fatigue results of all adjacent free surfaces are less than the fatigue result of the current free surface, the current free surface is marked as Peak; if the fatigue results of adjacent free surfaces are both less than and greater than the fatigue result of the current free surface, the current free surface is marked as Ignore; if the fatigue results of adjacent free surfaces are both less than and equal to the fatigue result of the current free surface, the current free surface is marked as Pending; if the fatigue results of adjacent free surfaces are both greater than and equal to the fatigue result of the current free surface, the current free surface is marked as Pending; if the current free surface is marked as Pending, a recursive method is used to continue checking neighboring nodes that are equal to it, until it is determined that the marking is not Pending, at which point the recursion ends.

[0027] This application also discloses a fatigue hotspot prediction system, including:

[0028] Memory, used to store computer-executable instructions; and,

[0029] A processor for implementing the steps of the method as described above when executing the computer-executable instructions.

[0030] The embodiments of this application can reduce testing difficulty, improve testing accuracy, and reduce testing costs, thereby significantly improving the efficiency of durability fatigue testing.

[0031] The specification of this application contains numerous technical features distributed across various technical solutions. Listing all possible combinations of these technical features (i.e., technical solutions) would make the specification excessively lengthy. To avoid this problem, the various technical features disclosed in the above-described invention, the various technical features disclosed in the following embodiments and examples, and the various technical features disclosed in the accompanying drawings can be freely combined to form various new technical solutions (all of which are considered to have been described in this specification), unless such a combination of technical features is technically infeasible. For example, one example discloses feature A+B+C, and another example discloses feature A+B+D+E. Features C and D are equivalent technical means that serve the same function, and technically only one needs to be used; they cannot be used simultaneously. Feature E can technically be combined with feature C. Therefore, the solution A+B+C+D should not be considered as described because it is technically infeasible, while the solution A+B+C+E should be considered as described. Attached Figure Description

[0032] Figure 1 This is a flowchart illustrating the fatigue hotspot prediction method according to the first embodiment of this application;

[0033] Figure 2 This is a schematic diagram of the fatigue hotspot prediction system according to the second embodiment of this application;

[0034] Figure 3 This is a schematic diagram of a virtual patch in an example of the fatigue hotspot prediction method according to the first embodiment of this application;

[0035] Figure 4 This is a schematic diagram of the overall coordinate system and the virtual measurement coordinate system in an example of the fatigue hotspot prediction method according to the first embodiment of this application;

[0036] Figure 5 This is a schematic diagram of a first-order hexahedral solid element model;

[0037] Figure 6 This is a schematic diagram of the node coordinate system and patch coordinate system of the product structure;

[0038] Figure 7 This is a schematic diagram of the patch coordinate system and the element coordinate system of the solid element;

[0039] Figure 8 This is a schematic diagram of the corner points and nodes of adjacent elements in the entire model, which is composed of several finite elements;

[0040] Figure 9 This is a schematic diagram of searching for fatigue hotspots in the fatigue hotspot prediction method according to the first embodiment of this application;

[0041] Figure 10 This is another schematic diagram of searching for fatigue hotspots in the fatigue hotspot prediction method according to the first embodiment of this application. Detailed Implementation

[0042] In the following description, many technical details are presented to help the reader better understand this application. However, those skilled in the art will understand that the technical solutions claimed in this application can be implemented even without these technical details and various variations and modifications based on the following embodiments.

[0043] Explanation of some concepts:

[0044] A free surface is an element face that is connected to only one element. In other words, if a free surface belongs to only one element, then it is a free surface. For example, if two elements are connected by a free surface, then the free surface belongs to both elements and is therefore not a free surface. All other free surfaces belong to only one element and are therefore free surfaces.

[0045] Free surface mesh data: This data contains the numbers of all free surfaces and their nodes. The free surface number is represented by its cell number and unit surface number together. For example, "N1-N6-N5-N2" is a free surface, and its number is E1-F1. The data of a free surface is represented as: E1-F1: N1-N6-N5-N2. The set of data for all free surfaces is the free surface mesh data.

[0046] The following is a brief summary of some of the innovative aspects of this application:

[0047] Based on the finite element method (FEM) results, all fatigue hotspots on the entire outer surface of the model are calculated. These hotspots comprehensively and intuitively reflect the location and direction of the most vulnerable areas for damage and crack initiation on the entire outer surface of the model. Then, based on this hotspot information, a rapid and accurate fatigue testing plan for the physical prototype can be formulated. Strain gauges are then attached according to the location and direction of these hotspots. The embodiments in this specification directly provide the attachment positions and directions of strain gauges that conform to the requirements of real-world fatigue testing of physical prototypes, greatly reducing the difficulty of test planning, improving test accuracy, and lowering experimental costs.

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0049] The first embodiment of this application relates to a fatigue hotspot prediction method. All products or components must undergo real fatigue testing after design to verify whether they meet the durability performance requirements. Structural cracks often initiate on the structural surface. Under complex load conditions, accurately identifying the location and orientation of strain gauges using the fatigue hotspot prediction technology based on fatigue results described in this specification is crucial. First, virtual strain gauges are applied to the entire outer surface of the model using virtual strain gauge technology based on fatigue analysis, and their fatigue damage or safety factor and orientation are calculated. Then, the fatigue hotspots are identified by comparing the fatigue results across the entire finite element model's outer surface mesh. During the actual fatigue test of the physical prototype, real strain gauges can be applied based on the location and orientation of these fatigue hotspots.

[0050] Specifically, the flow of the fatigue hotspot prediction method in this embodiment is as follows: Figure 1As shown, the method includes the following steps:

[0051] Step 110: Obtain the free surface mesh data of the finite element model of the mechanical structure.

[0052] The free surface of the entire model and the free surface mesh data composed of the nodes associated with the free surface are analyzed from the mesh data of the finite element model.

[0053] Specifically, the free surface mesh data includes the surface number and node number of each free surface in the model. For example, the surface number of a free surface can be "E1-F1", and the node number of the free surface can be "N1-N6-N5-N2".

[0054] Preferably, the surface number and node number of a free surface can be represented together as "E1-F1: N1-N6-N5-N2".

[0055] Step 120: Determine the fatigue result of each free surface based on the free surface mesh data.

[0056] Preferably, the fatigue result of each free surface can be the fatigue result of the center point of the free surface or the fatigue result of the node of the free surface.

[0057] Specifically, for fatigue at the center point of a free surface, a measurement coordinate system can be calculated based on the free surface mesh data, with the center point of the surface as the origin and the normal direction of the surface center as the z-axis. The finite element results (Tensor) can then be transformed into this coordinate system. Then, an appropriate fatigue algorithm can be selected to calculate and search for the worst-case fatigue result (Damage) or safety factor and its direction on the XY plane. The direction is represented by the angle between the X-axis of the measurement coordinate system and the target fatigue result (Damage or Safety Factor).

[0058] On the other hand, specifically for fatigue analysis of nodes on a free surface, a measurement coordinate system can be calculated based on the free surface mesh data, with the node location as the origin and the normal direction of the node location as the z-axis. The finite element results (Tensor) are then transformed to this coordinate system. An appropriate fatigue algorithm is then selected to calculate and search for the worst-case fatigue result (Damage) or safety factor and its direction on the XY plane. The direction is represented by the angle between the X-axis and the measurement coordinate system. It should be noted that because the z-axis normal of the measurement coordinate system is perpendicular to the tangent plane of the free surface center or the free node, the XY plane of the measurement coordinate system is also the tangent plane of the free surface center or the free node. Fatigue calculations are performed every 10 degrees on this tangent plane to find the direction of maximum damage. Specifically, for each free surface midpoint or free node, after establishing the measurement coordinate system, the stress / strain tensor at that location in this measurement coordinate system is calculated. Fatigue is calculated once, and the result Angle0:Damage0 is recorded. The measurement coordinate system is rotated 10 degrees around the Z-axis, and the stress / strain tensor at that location in this measurement coordinate system is calculated again. Fatigue is calculated again, and the result Angle10:Damage1 is recorded. This process continues until the fatigue result is obtained after rotating 170 degrees. From the 18 calculation results, the maximum Damage value and its corresponding angle are found. This value represents the most damaging value, and its direction represents the most damaging direction. If the Safety Factor is being calculated, the principle is similar, except that the minimum Safety Factor value and its direction are found later. This will not be elaborated upon here.

[0059] It should be noted that the free surface mesh data includes the free surface number and the node numbers that make up the free surface. The nodes that make up the free surface are called free nodes. The fatigue calculation location can be the center point of the free surface or on a free node. The shape of the free surface can be a triangle composed of 3 or 6 points and a quadrilateral composed of 4 or 8 points.

[0060] To better understand the technical solution of this application, a specific example is provided below. The details listed in this example are mainly for ease of understanding and are not intended to limit the scope of protection of this application.

[0061] The concepts involved in the following examples include:

[0062] Nonshared element face, also known as a non-shared element face, refers to the face contained by all elements in a model. Element faces exposed on the surface of the model are called "non-shared element faces." In addition, element faces shared by several elements and hidden inside the model are called "shared element faces." In the text, "non-shared element face," "target element face," and "free surface" have the same meaning and can be used interchangeably.

[0063] Element corner, also known as "element corner" in English, see [link / reference]. Figure 8 .

[0064] In a model composed of several finite elements, a node is defined as the corner point of adjacent elements that coincide. In other words, a node is called a "node" for the entire model, but a corner point for each individual element. See [link to relevant documentation]. Figure 8 .

[0065] The stress / strain tensor is a mathematical representation of stress / strain states. The " / " symbol here represents "and or".

[0066] The center of a face, also known as the "center of face," refers to the geometric center of a face.

[0067] Damage, in English, is a physical quantity that corresponds to lifespan; the greater the damage, the shorter the lifespan.

[0068] Measuring coordinate systems are reference systems used to determine strain / strain tensors.

[0069] Fatigue, in English, refers to the failure of a structure under repeated stress / strain. Life is commonly used to measure the fatigue durability of a structure.

[0070] In the following examples, the inventors, through long-term research, discovered that different locations within a structure experience varying stress and strain values ​​due to different stress states. The durability performance of a structure (fatigue life, cumulative damage, and fatigue safety factor) is extremely sensitive to its stress and strain state. Furthermore, the material strain gauge (SN) curves used to calculate the fatigue durability performance of a structure are obtained from strain gauge measurements attached to the structural surface. Stress and strain obtained from experimental measurements can be directly used for fatigue durability performance calculations; however, stress / strain tensors obtained from finite element analysis, especially for complex structures and stress states, are often unsuitable for direct use in fatigue algorithms to calculate fatigue durability performance. This is because directly using stress and strain from finite element analysis can lead to speed and accuracy issues in fatigue durability performance analysis. The reasons are as follows:

[0071] 1) The coordinate plane of the finite element node (analogous to a strain gauge) is not easily aligned with the tangent plane of the structural surface. The stress / strain tensor of the finite element node is often not the stress / strain value measured by the strain gauge on the structural surface, leading to further accuracy issues in fatigue durability calculations. 2) The stress and strain of a solid element are the stress and strain values ​​at the center of the solid element. Since the center of a solid finite element is not on the surface, the stress and strain at the center of the solid element often differ from the stress and strain measured at the structural surface, leading to further accuracy issues in fatigue durability calculations. 3) Solid elements inside the structure, since they are not located at the crack initiation surface, cannot represent the actual lifespan of the structure when calculated for internal solid elements, and this also incurs significant computational costs. 4) The coordinate plane of the finite element corner points is not easily aligned with the tangent plane of the structural surface. The stress / strain tensor of the finite element corner points is often not the stress / strain value measured by the surface strain gauge, leading to further accuracy issues in fatigue durability calculations. 5) The stress and strain values ​​of shared corner points of several finite elements are often different. If fatigue durability calculations are performed for the corner points of each element, the computational load is too large and the calculation speed is slow.

[0072] Specifically, such as Figure 5 As shown, taking a first-order hexahedral solid element model (solid element without intermediate nodes) as an example, with n1 meshes in the length direction, n2 meshes in the width direction, and n3 meshes in the height direction, the relationship between the number of non-shared element faces, the total number of faces, the number of elements, the number of nodes, and the number of corner points in this model and n1, n2, and n3 is as follows:

[0073] Number of non-shared unit faces = 2 × (n1 × n2 + n2 × n3 + n1 × n3)

[0074] The total number of faces = n1×n2 + n2×n3 + n1×n3 + 3×n1×n2×n3

[0075] Number of units = n1 × n2 × n3

[0076] Number of nodes = 4 × (n1+1) + 2 × (n1+1) × (n2-1) + (n3-1) × [2 × (n1+1) + (n1+1) × (n2-1)]

[0077] Number of corner points = 8 × n1 × n2 × n3

[0078] Further details are shown in the table below:

[0079]

[0080] The table above shows that as the size of the finite element increases, the difference between the number of non-shared element faces and the total number of faces in the element increases, which means that the computational cost of using the method based on non-shared element faces is more advantageous.

[0081] Furthermore, such as Figure 6 As shown, the finite element nodal coordinate system plane (analogous to strain gauges) is not easily aligned with the tangent plane of the structural surface. The stress / strain tensors of the finite element nodules are often not the stress-strain values ​​measured by strain gauges on the structural surface, leading to further accuracy issues in fatigue durability performance calculations. Figure 6 The left figure shows the nodal coordinate system. Figure 6 The right figure shows the patch coordinate system.

[0082] The structural shapes of actual products vary greatly, making it difficult to use perfectly regular meshes in practical calculations. This is especially true for structures whose surfaces are not entirely planar. The patch coordinate system is located on the tangent plane of the structural surface, while the nodal coordinate system is located on the nodes. Furthermore, the axes of these coordinate systems are often not parallel or perpendicular to the structural surface. This difference in axis direction leads to significant variations in the measured strain. Taking the angular difference of a single axis as an example, the table below illustrates the relative strain error caused by the angular difference:

[0083]

[0084]

[0085] Strain error can lead to even larger-scale errors in fatigue durability. The table below shows the life changes corresponding to strain variations, which can be seen that a 10% strain error can lead to several times the life error.

[0086] strain life 0.00010 1.287E+17 0.00011 4.720E+16 0.00012 1.889E+16 0.00013 8.134E+15 0.00014 3.728E+15 0.00015 1.803E+15

[0087] The stress and strain of a solid element are the stress and strain values ​​at the center of the solid element. Since the center of a solid finite element is not on the surface, there will often be a difference between the stress and strain at the center of the solid element and the stress and strain measured on the surface of the structure, which will lead to further problems in the accuracy of fatigue durability calculation.

[0088] For example Figure 7 Taking the solid unit shown as an example, Figure 7 The left image shows the patch coordinate system, and the right image shows the element coordinate system. The element coordinate system is located inside the element and measures at a different location than the patch coordinate system. The directions of the two coordinate systems are also different, so there will be errors in the measured strain / stress, which will lead to greater errors in the subsequent fatigue life results.

[0089] Figure 8The left image shows the patch coordinate system, and the right image shows the corner coordinate system. The stress and strain values ​​at shared corner points of several finite elements are often different. Performing fatigue durability calculations for each corner point of every element would be computationally intensive and slow. Furthermore, the coordinate plane of the finite element corner points is not easily aligned with the tangent plane of the structural surface. The stress / strain tensors at the finite element corner points are often not the stress and strain values ​​measured by surface strain gauges, leading to further accuracy issues in fatigue durability calculations.

[0090] In a model composed of several finite elements, the corner points of adjacent elements coincide and are called nodes. That is, they are called nodes for the entire model, but corner points for each individual element. One node corresponds to several corner points, so the computational cost of calculating based on corner points is several times (at least twice) that of calculating based on nodes.

[0091] Figure 8 The difference between the patch coordinate system and the corner coordinate system leads to a discrepancy between the strain / stress measured in the corner coordinate system and the strain / stress measured in the patch coordinate system, thus introducing greater errors into further fatigue calculations.

[0092] Based on the above analysis, in the following example, considering the engineering phenomenon that structural cracks often originate on the structural surface, the outer surface of the finite element model is first identified as the patch area using "non-shared element surface identification technology." Then, the patch position is determined using "virtual patch positioning technology," and "virtual strain gauges" are automatically applied to the outer surface of the finite element model. A virtual "measurement coordinate system" is established, and the mapping of finite element stress / strain tensors is achieved through "virtual measurement technology." The stress and strain results of finite element nodes on the non-shared node surface are converted into the stress and strain of the "virtual patch" position and corresponding direction under the "measurement coordinate system" used for fatigue calculation. This allows the calculation of structural fatigue durability to be performed on a surface consistent with material testing. The "virtual patch technology" for fatigue analysis can greatly reduce unnecessary calculations of internal structural elements, improving both calculation accuracy and speed.

[0093] The following examples demonstrate a non-shared unit surface identification technique. Specifically, based on searching for surfaces where fatigue may occur, this technique significantly reduces computational cost compared to traditional methods that calculate for all units or nodes of the entire model.

[0094] Furthermore, the following example proposes a virtual patch positioning technique. Specifically, virtual patches are applied to non-shared cell surfaces to determine a virtual measurement coordinate system, thereby obtaining patch positions and orientations consistent with the test and improving the consistency between calculation and testing.

[0095] Furthermore, the following examples demonstrate virtual measurement techniques to obtain stress and strain consistent with those measured, which helps improve the accuracy of fatigue durability analysis.

[0096] The following example illustrates a virtual patch measurement method for fatigue analysis of mechanical structures, which includes the following steps:

[0097] Based on the free surface mesh data, obtain the target element surface corresponding to each virtual patch in the finite element model of the mechanical structure;

[0098] Determine the position of the center of each target element surface in the first coordinate system and the stress / strain tensor matrix of the center of each target element surface relative to the first coordinate system;

[0099] A corresponding second coordinate system is established with the center of each target unit surface. The angle cosine matrix between the second coordinate system and each axis of the first coordinate system of the target unit surface is determined. Based on the angle cosine matrix between the second coordinate system and each axis of the first coordinate system of the target unit surface, and the stress / strain tensor matrix of the center of the target unit surface relative to the first coordinate system, the stress / strain tensor matrix of the center of the target unit surface relative to the second coordinate system is determined.

[0100] Based on the stress / strain tensor matrix of the center of each target element surface in the corresponding second coordinate system, the fatigue durability, and / or damage, and / or safety factor, and the direction of maximum damage of the virtual patch corresponding to the target element surface are determined, serving as the fatigue result of the free surface corresponding to the target element surface. Specifically:

[0101] First, obtain the element number of each element surface in the finite element model of the mechanical structure and the element number of all elements to which that element surface belongs.

[0102] Specifically, the unit surface can be a shared unit surface or a non-shared unit surface.

[0103] The element surface refers to the surface contained in all elements of the finite element model, including element surfaces exposed on the surface of the finite element model, called "non-shared element surfaces", and element surfaces shared by several elements and hidden inside the finite element model, called "shared element surfaces".

[0104] Each unit has a unique and fixed unit number.

[0105] Each element consists of several element faces; for example, a tetrahedral element has 4 element faces, and a hexahedral element has 6 element faces.

[0106] Preferably, the total number of elements to which a given element belongs can be determined by iterating through the number of element numbers corresponding to the elements to which each element belongs.

[0107] Then, each cell surface with a unique corresponding cell number is taken as the target cell surface corresponding to the virtual patch.

[0108] Specifically, determine whether the cell number corresponding to all cells to which each cell surface belongs obtained in the above steps is unique. If so, determine the cell surface as a non-shared cell surface and use it as the target cell surface of the virtual patch.

[0109] For example, if it is a hexahedral element with 6 element faces, one of which is an element face exposed on the surface of the finite element model, called a "non-shared element face", and the other 5 are element faces shared by several elements and hidden inside the finite element model, called "shared element faces".

[0110] When identifying the target cell surface of a virtual patch, the determination is made based on whether the cell number corresponding to all cells to which each cell surface belongs is unique.

[0111] It should be noted that the applicant of this application has found through research that, since the shared unit surface is located inside the mechanical mechanism, and fatigue crack initiation often starts from the surface, calculating the fatigue life of the shared unit surface inside the structure (its life value is greater than that of the structural surface) is not valuable for judging the minimum life of the structure, but will instead cause an excessive amount of calculation.

[0112] Then, the position of the center of each target element surface in the global coordinate system and the stress / strain tensor matrix relative to the global coordinate system are determined.

[0113] Preferably, the stress / strain tensor of each corner point of each target element surface relative to the global coordinate system can be determined first, and then the stress / strain tensor matrix of the center of the target element surface relative to the global coordinate system can be determined based on the stress / strain tensor of each corner point relative to the global coordinate system.

[0114] Specifically, the normal direction r, corner points, and stress / strain tensor of each target element surface are read.

[0115] See Figure 3 The hexahedral element Element-1 has two virtual patch target element faces: target element face A1, which is composed of nodes C1-C2-C3-C4, and target element face A2, which is composed of nodes C5-C6-C2-C1. The normal directions of target element face A1 and target element face A2 are set to r1 and r2, respectively (not shown in the figure).

[0116] The following example, using target cell surface A1, further explains the specific implementation method for locating the target cell surface of the virtual patch:

[0117] Determine the number of nodes on target element surface A1, NA1 = 4; the coordinates of each node on target element surface A1 in the global coordinate system o-xyz, C1(x_C1,y_C1,z_C1), C2(x_C2,y_C2,z_C2), C3(x_C3,y_C3,z_C3), and C4(x_C4,y_C4,z_C4); and the normal direction r1 of target element surface A1. Also determine the stress / strain tensor s of each node relative to the global coordinate system o-xyz. Taking node C1 as an example, the stress / strain tensor of node C1 in the global coordinate system is:

[0118] s_c1(sxx_c1,syy_c1,szz_c1,sxy_c1,syz_c1,sxz_c1)

[0119] in,

[0120] sxx_c1, syy_c1, and szz_c1 represent the normal stress / strain of point C1 in the x, y, and z directions, respectively, in the global coordinate system o-xyz.

[0121] sxy_c1, syz_c1, and sxz_c1 represent the shear stress / strain at this point in the xy, yz, and xz planes, respectively.

[0122] It should be noted that the overall coordinate system o-xyz is a global coordinate system, which is the system reference relative to the overall model of the mechanical structure.

[0123] In this embodiment, the global coordinate system can be referred to as the first coordinate system.

[0124] Based on the coordinate values ​​of each node (node ​​C1, node C2, node C3, node C4) obtained from statistics, the coordinates oA1(x_oA1, y_oA1, z_oA1) of the center position oA1 of the target unit surface of the virtual patch are determined according to the following formula:

[0125]

[0126]

[0127]

[0128] Furthermore, based on the stress / strain tensor of each corner point of the target element surface A1 relative to the global coordinate system, the stress / strain tensor of the center point oA1 of the target element surface A1 in the global coordinate system is determined (represented uniformly as s_oA1(sxx_oA1,syy_oA1,szz_oA1,sxy_oA1,syz_oA1,sxz_oA1)).

[0129] Taking the normal stress / strain tensor sxx_oA1 of the center point oA1 of the target element surface A1 in the x-direction of the global coordinate system o-xyz as an example:

[0130]

[0131] Using the same method, the stress / strain tensor matrix [s_oA1] of the center of the target element surface relative to the global coordinate system can be determined based on the stress / strain tensor of each corner point relative to the global coordinate system.

[0132] Wherein, sxx_oA1, syy_oA1, and szz_oA1 are the normal stresses / strains of point oA1 in the x, y, and z directions of the global coordinate system o-xyz, respectively; sxy_oA1, syz_oA1, and sxz_oA1 are the shear stresses / strains of the point in the xy, yz, and xz planes, respectively.

[0133] Then, a corresponding virtual measurement coordinate system is established with the center of each target unit surface. The angle cosine matrix between the virtual measurement coordinate system of the target unit surface and each axis of the global coordinate system is determined. Based on the angle cosine matrix between the virtual measurement coordinate system of the target unit surface and each axis of the global coordinate system, and the stress / strain tensor matrix of the center of the target unit surface relative to the global coordinate system, the stress / strain tensor matrix of the center of the target unit surface relative to the virtual measurement coordinate system is determined.

[0134] Preferably, the angle cosine matrix between each axis of the virtual measurement coordinate system and the global coordinate system refers to the angle cosine matrix between each axis of the virtual measurement coordinate system and the global coordinate system with the center of the target unit surface as the origin.

[0135] Preferably, a virtual measurement coordinate system can be established at the center of the target unit surface of the virtual patch in the following specific manner:

[0136] 1) The virtual measurement coordinate axis oA1-zA1 is defined as the axis parallel to the normal r1 of the target element surface A1 passing through the center point oA1 and pointing in its positive direction. 2) The projection of the line connecting node C1 and node C2 onto the plane passing through the center point oA1 and perpendicular to the normal r1 of the target element surface A1 is defined as the virtual measurement coordinate axis oA1-xA1. It should be noted that C1 and C2 are two points on the element surface, and there is no specific requirement for their association. The virtual measurement coordinate axis oA1-xA1 lies on a plane perpendicular to the virtual measurement coordinate axis oA1-zA1, and its direction is the projection of the ray from node C1 to node C2 onto this plane. 3) Based on oA1-xA1 and oA1-zA1, the virtual measurement coordinate axis oA1-yA1 is determined using the right-hand rule.

[0137] In this embodiment, the virtual measurement coordinate system can be referred to as the second coordinate system.

[0138] See Figure 4 In this step, the angle cosine matrix [L] between each axis of the virtual measurement coordinate system with the center oA1 of the target unit surface as the origin and each axis of the global coordinate system is determined.

[0139] The stress / strain tensor matrix s′_oA1 of the center oA1 of the target element surface in the virtual measurement coordinate system oA1-xA1,yA1,zA1 is:

[0140]

[0141] Then, based on the stress / strain tensor matrix of the center of each target unit surface in the corresponding virtual measurement coordinate system, the fatigue durability life, and / or damage, and / or safety factor of the virtual patch corresponding to the target unit surface are determined.

[0142] Preferably, the stress / strain tensor matrix of the center of each target unit surface in the corresponding virtual measurement coordinate system at different times can be obtained first to determine the amplitude and mean of the time history of the stress / strain tensor matrix of the center of the target unit surface in the corresponding virtual measurement coordinate system. Then, a fatigue algorithm is used to calculate the fatigue durability life, and / or damage, and / or safety factor of the center position of the virtual patch corresponding to each target unit surface.

[0143] The following example, using alternating stress cycles, illustrates the process of using the stress / strain tensor obtained from virtual measurements for biaxial Manson-Coffin strain fatigue calculations:

[0144]

[0145] Assume s′yy oA1 <s′xx oA1First, calculate the biaxial ratio r and the strain amplitude Δε:

[0146]

[0147]

[0148] Then, the biaxial ratio r and strain amplitude Δε are substituted into the biaxial Manson-Coffin strain lifetime calculation formula, and the lifetime N is calculated through nonlinear solution. f :

[0149]

[0150] Among them, Young's modulus E, fatigue strength coefficient σ f Poisson's ratio μ, fatigue strength index b, fatigue ductility coefficient ε f Both ′ and fatigue ductility index c are material parameters corresponding to the structure.

[0151] The stress / strain tensor obtained from virtual measurement can be used in various fatigue algorithms. When using other fatigue algorithms to calculate life / damage or fatigue safety factor, other terms of the s′_oA1 matrix may be used, which will not be elaborated here.

[0152] In the above example, based on the engineering phenomenon that structural cracks often originate on the surface of the structure, the outer surface of the finite element model is first identified as the patch area using "non-shared element surface identification technology". Then, the patch position is determined using "virtual patch positioning technology". "Virtual strain gauges" are automatically applied to the outer surface of the finite element model, and a virtual "measurement coordinate system" is established. Through "virtual measurement technology", the finite element stress / strain tensor is mapped, and the stress and strain results of the finite element nodes on the non-shared node surface are converted into the stress and strain of the "virtual patch" position and corresponding direction under the "measurement coordinate system" for fatigue calculation. The calculation of the structural fatigue durability performance is returned to the surface consistent with material testing. The "virtual patch technology" for fatigue analysis can greatly reduce the unnecessary calculation of the internal elements of the structure, not only improving the calculation accuracy but also increasing the calculation speed.

[0153] In the above examples, a non-shared unit surface identification technique was proposed. Specifically, based on searching for surfaces where fatigue may occur, this technique significantly reduces the computational load compared to traditional methods that calculate for all units or nodes of the entire model.

[0154] Furthermore, in the above examples, a virtual patch positioning technology was proposed. Specifically, virtual patches were placed on the non-shared unit surface to determine a virtual measurement coordinate system, obtain patch positions and orientations consistent with the test, and improve the consistency between calculation and test.

[0155] Furthermore, in the above examples, virtual measurement technology was proposed to obtain stress and strain consistent with the test, which is beneficial to improving the accuracy of fatigue durability analysis.

[0156] Step 130: Search for fatigue hotspots in the finite element model of the mechanical structure based on the fatigue results of each free surface.

[0157] Specifically, if the fatigue at the center point of a free surface is being evaluated, the process first iterates through the entire free surface grid to find the free surface with the largest fatigue result at its center point and sets it as the current free surface. Then, starting from the current free surface, the fatigue at the center point is compared with that of its adjacent free surfaces, and these comparisons are marked. This process continues until all free surfaces have been traversed. If the fatigue result being compared is Damage, all points marked as Peak are considered fatigue hotspots. If the fatigue result being compared is Safety factor, all points marked as Valley are considered fatigue hotspots. This means that a complete fatigue hotspot prediction either compares and searches for hotspots based on the fatigue result damage value or the fatigue result safety factor value. If the comparison is based on damage value, the point with the largest damage value is the hotspot; if the comparison is based on safety factor value, the point with the smallest safety factor value is the hotspot.

[0158] Specifically, the comparison and labeling methods can be implemented in the following ways:

[0159] 1) If the fatigue results of all adjacent free surfaces are greater than the fatigue result of the current free surface, then mark the current free surface as Valley.

[0160] 2) If the fatigue results of all adjacent free surfaces are less than the fatigue result of the current free surface, then mark the current free surface as Peak.

[0161] 3) If the fatigue results of adjacent free surfaces are both less than or greater than the fatigue results of the current free surface, then mark the current free surface as Ignore.

[0162] 4) If the fatigue result of an adjacent free surface is both less than or equal to the fatigue result of the current free surface, then mark the current free surface as Pending. If the fatigue result of an adjacent free surface is both greater than or equal to the fatigue result of the current free surface, then mark the current free surface as Pending. If the current free surface is marked as Pending, a recursive method is needed to continue checking neighboring nodes that are equal to it, until it is determined that the mark is not Pending, at which point the recursion ends. See also Figure 9 .

[0163] like Figure 9 As shown, to determine whether the point with a value of 9 on the left is a peak, we need to further determine whether the point with a value of 9 on the right is larger than the values ​​of all the points around it. If they are all larger, then both points with a value of 9 are peak points. If the value of any of the points around the point with a value of 9 on the right is larger than it, then both points with a value of 9 should be marked as Ignore to exclude them.

[0164] It is understandable that the method for evaluating the nodes of a free surface is the same as the method for evaluating the center point of a free surface, such as... Figure 10 As shown, Figure 10 The left image shows the current free surface and its adjacent free surfaces, while the right image shows the current free surface node and its adjacent free surface nodes.

[0165] Step 140: Output the fatigue hotspots of the finite element model of the mechanical structure.

[0166] The above embodiments can reduce testing difficulty, improve testing accuracy, and reduce testing costs, thereby significantly improving the efficiency of durability fatigue testing.

[0167] The second embodiment of this application relates to a fatigue hotspot prediction system, the structure of which is as follows: Figure 2 As shown, the fatigue hotspot prediction system includes:

[0168] The free surface network data acquisition module is used to acquire the free surface mesh data of the finite element model of the mechanical structure;

[0169] The fatigue result acquisition module is used to determine the fatigue result of each free surface based on the free surface mesh data;

[0170] The fatigue result search module is used to search for fatigue hotspots in the finite element model of the mechanical structure based on the fatigue results of each free surface.

[0171] The output module is used to output the fatigue hotspots of the finite element model of the mechanical structure.

[0172] Preferably, when the fatigue result search module searches for fatigue hotspots in the finite element model of the mechanical structure based on the fatigue results of each free surface, if the fatigue hotspots are searched by comparing the fatigue result damage values, the point with the largest damage value is the fatigue hotspot; if the fatigue hotspots are searched by comparing the fatigue result safety factor values, the point with the smallest safety factor value is the fatigue hotspot.

[0173] Preferably, when the fatigue result search module searches for fatigue hotspots in the finite element model of the mechanical structure based on the fatigue results of each free surface, if the fatigue results of all adjacent free surfaces are greater than the fatigue result of the current free surface, the current free surface is marked as Valley; if the fatigue results of all adjacent free surfaces are less than the fatigue result of the current free surface, the current free surface is marked as Peak; if the fatigue results of adjacent free surfaces are both less than and greater than the fatigue result of the current free surface, the current free surface is marked as Ignore; if the fatigue results of adjacent free surfaces are both less than and equal to the fatigue result of the current free surface, the current free surface is marked as Pending; if the fatigue results of adjacent free surfaces are both greater than and equal to the fatigue result of the current free surface, the current free surface is marked as Pending; if the current free surface is marked as Pending, a recursive method is used to continue checking neighboring nodes that are equal to it, until it is determined that the marking is not Pending, at which point the recursion ends.

[0174] The first embodiment is a method embodiment corresponding to this embodiment. The technical details in the first embodiment can be applied to this embodiment, and the technical details in this embodiment can also be applied to the first embodiment.

[0175] It should be noted that those skilled in the art should understand that the functions of each module shown in the above-described fatigue hotspot prediction system implementation can be understood with reference to the relevant description of the aforementioned fatigue hotspot prediction method. The functions of each module shown in the above-described fatigue hotspot prediction system implementation can be implemented by a program (executable instructions) running on a processor, or by specific logic circuits. If the fatigue hotspot prediction system described in this application is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application embodiment, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, mobile hard drive, read-only memory (ROM), magnetic disk, or optical disk. Thus, this application embodiment is not limited to any specific hardware and software combination.

[0176] Accordingly, this application also provides a computer storage medium storing computer-executable instructions, which, when executed by a processor, implement the various method implementations of this application.

[0177] Furthermore, this application also provides a fatigue hotspot prediction system, including a memory for storing computer-executable instructions and a processor; the processor is used to implement the steps in the above-described method embodiments when executing the computer-executable instructions in the memory. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The aforementioned memory may be read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or solid-state drive, etc. The steps of the methods disclosed in the various embodiments of this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor.

[0178] It should be noted that in this patent application, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. In this patent application, if it refers to performing an action according to an element, it means performing the action at least according to that element, including two cases: performing the action only according to that element, and performing the action according to that element and other elements. Expressions such as "multiple," "repeatedly," and "various" include two, two times, two kinds, and more than two, more than two times, and more than two kinds.

[0179] All documents mentioned in this application are considered to be incorporated in their entirety into the disclosure of this application so that they can serve as a basis for modifications if necessary. Furthermore, it should be understood that after reading the foregoing disclosure of this application, those skilled in the art can make various alterations or modifications to this application, and these equivalent forms also fall within the scope of protection claimed in this application.

Claims

1. A method for predicting fatigue hotspots, characterized in that, include: Obtain the free surface mesh data of the finite element model of the mechanical structure; The fatigue result for each free surface is determined based on the free surface mesh data; Based on the fatigue results of each free surface, the fatigue hotspots of the finite element model of the mechanical structure are searched. The search includes: traversing the entire free surface mesh, comparing the fatigue results of the current free surface with its adjacent free surfaces, and marking them accordingly: if the fatigue results of all adjacent free surfaces are greater than the fatigue result of the current free surface, the current free surface is marked as Valley; if the fatigue results of all adjacent free surfaces are less than the fatigue result of the current free surface, the current free surface is marked as Peak; if the fatigue results of adjacent free surfaces are both less than and greater than the fatigue result of the current free surface, the current free surface is marked as Ignore; if the fatigue results of adjacent free surfaces are... If the fatigue result of the current free surface is both less than or equal to the fatigue result of the current free surface, then the current free surface is marked as Pending. If the fatigue result of an adjacent free surface is both greater than or equal to the fatigue result of the current free surface, then the current free surface is marked as Pending. If the current free surface is marked as Pending, a recursive method is used to continue checking neighboring nodes that are equal to it until it is determined that the mark is not Pending, at which point the recursion ends. If the fatigue hotspots are searched by comparing the fatigue result damage value, then all free surfaces marked as Peak are fatigue hotspots. If the fatigue hotspots are searched by comparing the fatigue result safety factor value, then all free surfaces marked as Valley are fatigue hotspots. Output the fatigue hotspots of the finite element model of the mechanical structure.

2. The method as described in claim 1, characterized in that, The free surface refers to the element surface exposed on the surface of the finite element model.

3. The method as described in claim 1, characterized in that, The fatigue result of the free surface is the fatigue result of the center point of the free surface, or the fatigue result of the node of the free surface.

4. The method as described in claim 1, characterized in that, Determining the fatigue result of each free surface based on the free surface mesh data includes the following sub-steps: Based on the free surface mesh data, obtain the target element surface corresponding to each virtual patch in the finite element model of the mechanical structure; Determine the position of the center of each target element surface in the first coordinate system and the stress / strain tensor matrix of the center of each target element surface relative to the first coordinate system; A corresponding second coordinate system is established with the center of each target unit surface. The angle cosine matrix between the second coordinate system and the first coordinate system of the target unit surface is determined. Based on the angle cosine matrix between the second coordinate system and the first coordinate system of the target unit surface, and the stress / strain tensor matrix of the center of the target unit surface relative to the first coordinate system, the stress / strain tensor matrix of the center of the target unit surface relative to the second coordinate system is determined. as well as Based on the stress / strain tensor matrix of the center of each target unit surface in the corresponding second coordinate system, the fatigue durability life, and / or damage, and / or safety factor and the most damaged direction of the virtual patch corresponding to the target unit surface are determined as the fatigue result of the free surface corresponding to the target unit surface.

5. A fatigue hotspot prediction system, characterized in that, Include: The free surface network data acquisition module is used to acquire the free surface mesh data of the finite element model of the mechanical structure; The fatigue result acquisition module is used to determine the fatigue result of each free surface based on the free surface mesh data; The fatigue result search module is used to search for fatigue hotspots in the finite element model of the mechanical structure based on the fatigue results of each free surface. The module traverses the entire free surface mesh, comparing the fatigue results of the current free surface with its adjacent free surfaces and marking them accordingly: if the fatigue results of all adjacent free surfaces are greater than the current free surface's fatigue result, the current free surface is marked as Valley; if the fatigue results of all adjacent free surfaces are less than the current free surface's fatigue result, the current free surface is marked as Peak; if the fatigue results of adjacent free surfaces are both less than and greater than the current free surface's fatigue result, the current free surface is marked as Ignore; if... If the fatigue result of an adjacent free surface is both less than or equal to the fatigue result of the current free surface, then the current free surface is marked as Pending. If the fatigue result of an adjacent free surface is both greater than or equal to the fatigue result of the current free surface, then the current free surface is marked as Pending. If the current free surface is marked as Pending, a recursive method is used to continue checking neighboring nodes that are equal to it until it is determined that the mark is not Pending, at which point the recursion ends. If the fatigue hotspots are searched by comparing the fatigue result damage value, then all free surfaces marked as Peak are fatigue hotspots. If the fatigue hotspots are searched by comparing the fatigue result safety factor value, then all free surfaces marked as Valley are fatigue hotspots. The output module is used to output the fatigue hotspots of the finite element model of the mechanical structure.

6. A fatigue hotspot prediction system, characterized in that, include: Memory is used to store executable instructions for a computer; as well as, A processor configured to implement the steps of the method as described in any one of claims 1 to 4 when executing the computer-executable instructions.

Images