Building structure analysis method based on multi-scale finite elements

Abstract

The application discloses a kind of building structure analysis methods based on multi-scale finite element, comprising the following steps: determining calculation target, including the whole rod shell calculation model to be analyzed is established, the part in building structure is divided into two analysis site types of micro unit site, macro unit site;Establish a geometric model to obtain the overall multi-scale geometric model;Establish analysis model;Model verification;Step five, result analysis and output.Compared with prior art, it solves the problem that the analysis result is not accurate due to the fact that the model boundary condition simulation is not real or the selected type and mesh size of the finite element of the key analysis component is rough in the traditional finite element analysis method, and the modeling work is too complex in the traditional method.This analysis method not only can simplify the overall calculation degree of freedom, improve the modeling and calculation efficiency, but also can refine the finite element type and mesh size of the key analysis component, and truly simulate the spatial stress state of the analysis component.

Description

Technical Field

[0001] This invention relates to a building technology, and more particularly to a multi-scale finite element-based building structure analysis method for the design and analysis of high-rise, super high-rise, and large-span structures. Background Technology

[0002] Traditional finite element analysis methods mainly fall into two categories. One approach involves first identifying key analytical nodes, then manually modeling and assigning physical properties to these nodes from the overall structure. This method uses relatively fine-grained finite element types and mesh sizes, and applies simple boundary constraints to the components, ultimately forming a small-scale finite element sub-model. The advantages of this traditional method are its small model scale, low degree of freedom at the element nodes, and fast computation speed. However, because the analytical nodes are isolated finite element sub-models, the boundary conditions of these sub-models are not accurately simulated, resulting in insufficient computational accuracy.

[0003] Another approach is to model the entire structure, considering all structural components. However, the type and mesh size of the finite elements for the components are relatively coarse, and the calculation results only observe the key analysis areas. Compared to the former, although this method considers the simplification of the overall calculation degrees of freedom and the real boundary conditions of the overall structure, it cannot well simulate the spatial stress state of the components because the type and mesh size of the finite elements selected for the key analysis areas are relatively coarse. Summary of the Invention

[0004] The purpose of this invention is to provide a building structure analysis method based on multi-scale finite element method. The technical problem to be solved is to simplify the overall calculation degree of freedom, improve modeling and calculation efficiency, and improve the accuracy of analysis results.

[0005] To solve the above problems, the present invention adopts the following technical solution: a building structure analysis method based on multi-scale finite element method, comprising the following steps:

[0006] Step 1: Determine the calculation objectives, including establishing a calculation model of the overall rod and shell to be analyzed, and dividing the parts in the building structure into two types of analysis parts: micro-unit parts and macro-unit parts;

[0007] Step 2: Establish a geometric model. Based on the geometric data in the overall rod and shell calculation model, perform three-dimensional modeling to obtain a three-dimensional overall rod and shell geometric model. Generate macroscopic and microscopic geometric components according to the type of analysis part, and process the interface between the two types of components to obtain a multi-scale geometric model of the whole.

[0008] Step 3: Establish the analysis model. A multi-scale finite element model is established using multi-scale geometric models, boundary contact geometric information, and overall rod-shell calculation model data.

[0009] Step 4: Model Verification. Determine the rationality of the overall multi-scale finite element analysis model. Calculate the period, inter-story displacement, and internal forces of the components in the micro-unit parts of the overall multi-scale finite element model and the overall rod and shell calculation model. If the difference between the period and inter-story displacement is less than 5%, and the internal forces of the components in the micro-unit parts are less than 10%, then the overall multi-scale finite element model is considered reasonable, and the verification of the overall multi-scale finite element model is completed. Otherwise, repeat steps 2 to 3 to fine-tune the geometric components and boundary contact geometry of the micro-unit parts until the overall multi-scale finite element model meets the above conditions.

[0010] Step 5: Analyze and output the results. Based on the reasonable overall multi-scale finite element model obtained in Step 4, determine the rationality of the structural system, components, and nodes. Rationality includes whether the overall structural indicators, component stiffness and strength, and stress-strain levels of the node units all meet the design objectives. If they meet the requirements, end the analysis and output the results. Otherwise, repeat Steps 2 to 4 to fine-tune the geometric components or nodes of the micro-units until the rationality conditions are met.

[0011] Furthermore, in step one, a complete rod-shell calculation model of the building structure to be analyzed is established. Based on the complexity of the components or nodes of the complete rod-shell calculation model and the nonlinearity of the component materials during loading, it is divided into two types of analysis parts: micro-unit parts and macro-unit parts.

[0012] Furthermore, nodes with more than 4 converging components or node domain size larger than 1.5 times the component cross-sectional size in terms of node complexity, and components and nodes with material stress exceeding yield strength or material strain exceeding ultimate strain in terms of nonlinearity are defined as micro-unit parts, and the remaining parts are macro-unit parts.

[0013] Furthermore, step two includes:

[0014] Three-dimensional modeling is performed based on the geometric data in the integral rod-shell calculation model to obtain the three-dimensional integral rod-shell geometric model.

[0015] Based on the type of analysis area, macroscopic and microscopic geometric components are generated, and the interface between the two types of components is processed to obtain a comprehensive multi-scale geometric model. Boundary contact geometric information is established and stored based on the boundary contact conditions of the macroscopic and microscopic geometric components.

[0016] Further, in step three, the analysis model is generated using a Python script in the commercial finite element analysis software ABAQUS.

[0017] Furthermore, the percentage of the difference in periods in step four is calculated using the following formula:

[0018] The percentage of the difference between periods = |T2 / T1-1|

[0019] Where T2 is the period of the overall multi-scale finite element model, and T1 is the period of the overall rod-shell calculation model.

[0020] Furthermore, the percentage of the difference in inter-story displacement in step four is calculated using the following formula:

[0021] The percentage of the difference in inter-story drift = |d2 / d1-1|

[0022] Where: d2 is the inter-story displacement of the overall multi-scale finite element model, and d1 is the inter-story displacement of the overall rod-shell calculation model.

[0023] Furthermore, the proportion of the internal force difference in the micro-unit parts of the components in step four is calculated using the following formula:

[0024] The percentage difference between the internal forces of the second component and the internal forces of the first component = |N2 / N1-1|

[0025] Where: N2 is the second component internal force in the micro-unit part of the overall multi-scale finite element model, and N1 is the first component internal force in the micro-unit part of the overall rod-shell calculation model.

[0026] Furthermore, the fine-tuning includes determining whether the ends of the macroscopic geometric components and the end faces of the microscopic geometric components are actually aligned; otherwise, an alignment operation is performed, and then the element type and mesh of the geometric components are reselected, as well as the boundary contact geometry type is selected.

[0027] Furthermore, the design objectives are based on the "Code for Design of Concrete" (GB50010-2010), the "Code for Seismic Design of Buildings" (GB50011-2010), and the "Technical Specification for Concrete Structures of High-Rise Buildings" (JGJ3-2010).

[0028] Compared with existing technologies, this invention solves the problems of inaccurate analysis results caused by unrealistic simulation of model boundary conditions or coarse selection of finite element types and mesh sizes for key components in traditional finite element analysis methods, as well as the excessive complexity of traditional modeling work. This analysis method not only simplifies the overall calculation of degrees of freedom and improves modeling and calculation efficiency, but also refines the finite element types and mesh sizes for key components, realistically simulating the spatial stress state of the components. Attached Figure Description

[0029] Figure 1 This is a flowchart of the analysis method of the present invention.

[0030] Figure 2 This is a schematic diagram of the west section elevation of the tower.

[0031] Figure 3 This is a schematic diagram of the structural plan of the floor (L49) where the tower's connecting trusses are located.

[0032] Figure 4 This is a schematic diagram of the non-connected floor (L41) structure of the tower.

[0033] Figure 5 This is a schematic diagram of the geometric model of the Rhino monolithic rod shell.

[0034] Figure 6 This is a schematic diagram of the overall multi-scale geometric model of Rhino (only the contiguous area is shown).

[0035] Figure 7 This is a schematic diagram of the geometric model of the Rhino truss.

[0036] Figure 8 This is a schematic diagram showing the contact between the end of the column member and the end face of the solid wall.

[0037] Figure 9 This is a schematic diagram showing the contact between the edge of the shell wall and the end face of the solid wall.

[0038] Figure 10 This is a schematic diagram of the contact between two solid end faces.

[0039] Figure 11 This is a schematic diagram showing the built-in contact between the solid wall and some truss component nodes.

[0040] Figure 12 This is a schematic diagram of the overall multi-scale finite element model established using ABAQUS.

[0041] Figure 13 This is a schematic diagram of the extent of the connected truss in a multi-scale finite element model.

[0042] Figure 14-1 A schematic diagram comparing the internal forces of the bracing members of the connected truss in a multi-scale finite element model and a monolithic rod-shell calculation model. Figure 1 .

[0043] Figure 14-2 A schematic diagram comparing the internal forces of the bracing members of the connected truss in a multi-scale finite element model and a monolithic rod-shell calculation model. Figure 2 .

[0044] Figure 14-3 A schematic diagram comparing the internal forces of the bracing members of the connected truss in a multi-scale finite element model and a monolithic rod-shell calculation model. Figure 3 .

[0045] Figure 15-1 Mises stress cloud of the conjoined truss node in the multi-scale finite element model of this invention. Figure 1 .

[0046] Figure 15-2 Mises stress cloud of the conjoined truss node in the multi-scale finite element model of this invention. Figure 2 .

[0047] Figure 15-3 Mises stress cloud of the conjoined truss node in the multi-scale finite element model of this invention. Figure 3 .

[0048] Figure 15-4 Mises stress cloud of the conjoined truss node in the multi-scale finite element model of this invention. Figure 4 .

[0049] Figure 15-5 Mises stress cloud of the conjoined truss node in the multi-scale finite element model of this invention. Figure 5 .

[0050] Figure 16-1 This invention relates to the multi-scale finite element model of the principal tensile and compressive stress cloud of adjacent walls in a connected truss structure. Figure 1 .

[0051] Figure 16-2 This invention relates to the multi-scale finite element model of the principal tensile and compressive stress cloud of adjacent walls in a connected truss structure. Figure 2 .

[0052] Figure 17-1 This is a schematic diagram of the connection nodes of the chords and web members in the overall multi-scale geometric model of the connected truss, which was modeled using Rhino 3D modeling software in this invention.

[0053] Figure 17-2 To Figure 17-1 A schematic diagram of the connection nodes between the chord and web members after fine-tuning.

[0054] Figure 17-3 To Figure 17-1 A schematic diagram of the connection nodes between the chord and web members after fine-tuning.

[0055] Figure 17-4 To Figure 17-1 A schematic diagram of the connection nodes between the chord and web members after fine-tuning.

[0056] Figure 17-5 To Figure 17-1 A schematic diagram of the connection nodes between the chord and web members after fine-tuning.

[0057] Figure 17-6 To Figure 17-1 A schematic diagram of the connection nodes between the chord and web members after fine-tuning.

[0058] Figure 18-1 for Figure 17-1 Mises stress contour plot.

[0059] Figure 18-2 for Figure 17-2 Mises stress contour plot.

[0060] Figure 18-3 for Figure 17-3 Mises stress contour plot.

[0061] Figure 18-4 for Figure 17-4 Mises stress contour plot.

[0062] Figure 18-5 for Figure 17-5 Mises stress contour plot.

[0063] Figure 18-6 for Figure 17-6 Mises stress contour plot. Detailed Implementation

[0064] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0065] like Figure 1 As shown, this invention discloses a building structure analysis method based on multi-scale finite element method, comprising the following steps:

[0066] Step 1: Determine the calculation objective

[0067] Defining the calculation objectives involves selecting a portion of the building structure as the first micro-unit and the remaining portion as the first macro-unit. Each micro-unit contains the internal forces of the first structural member. The overall frame-shell calculation model includes information such as period and inter-story displacement.

[0068] Specifically, an integral skeleton calculation model of the building structure to be analyzed is established. Existing structural design calculation software (such as YJK, PKPM, Structure Analysis Program 2000 (SAP2000), etc.) is used to input the information of the building structure to be analyzed to obtain the integral skeleton calculation model. Based on the complexity of the components or nodes of the integral skeleton calculation model and the degree of nonlinearity of the component materials during loading, it is divided into two types of analysis parts: micro-unit parts and macro-unit parts. In seismic design, some components in the entire building structure system are relatively important; these components are the main load-bearing components and are defined as key components. Key components must meet certain performance levels. Based on these design conditions, components that need to be focused on for analysis are selected. Among them, nodes with more than four intersecting components or node domain sizes larger than 1.5 times the component cross-sectional dimensions, and components and nodes with material stress exceeding yield strength or material strain exceeding ultimate strain in terms of nonlinearity, are defined as the first micro-unit parts. The remaining parts are the first macro-unit parts.

[0069] The integral frame-shell calculation model is established based on the architectural drawings, in which the walls are treated as shell units and the columns and beams as frame units, collectively referred to as the integral frame-shell calculation model.

[0070] Step 2: Establish a geometric model

[0071] Establishing a geometric model includes:

[0072] Using commercial software such as AutoCAD (CAD) or Rhino 3D modeling software, a 3D model is created based on the geometric data in the overall rod and shell calculation model. This results in a 3D overall rod and shell geometric model. Based on the analysis part type, macroscopic and microscopic geometric components are generated that correspond one-to-one with the first macroscopic unit part and the first microscopic unit part. The interface between the two types of geometric components is then processed to obtain the overall multi-scale geometric model. The geometric data includes the geometric data of the rod element and the shell element.

[0073] The three-dimensional geometric model of the integral rod and shell is obtained by performing three-dimensional modeling based on the data in the integral rod and shell calculation model using commercial software such as AutoCAD or Rhino 3D modeling software. Specifically, the geometric modeling interface (API interface) of AutoCAD or Rhino 3D modeling software is used to read the data of the integral rod and shell calculation model (YJK, PKPM, SAP2000, etc. calculation model data) and then generate the three-dimensional integral rod and shell geometric model using the existing AutoCAD or Rhino 3D modeling software.

[0074] Based on the analysis part type, macroscopic and microscopic geometric components are generated, and the interface between the two types of components is processed to obtain the overall multi-scale geometric model. Specifically, based on the first microscopic unit part defined in step one, geometric entities are generated for the rod and shell components in these first microscopic unit parts. The geometric entities include components or nodes in the microscopic unit parts. These geometric entities are defined as microscopic geometric components. All rod and shell components in the first macroscopic unit part are positioned as macroscopic geometric components. The interface between the two types of components is processed and optimized. The processing and optimization includes aligning the ends of the macroscopic geometric components with the ends of the microscopic geometric components and cutting the microscopic geometric components through Boolean operations in the prior art to obtain the overall multi-scale geometric model.

[0075] To ensure the continuity of the boundaries between macroscopic and microscopic geometric components, boundary contact geometry information is established and stored based on the boundary contact conditions of macroscopic and microscopic geometric components. The boundary contact geometry information includes boundary contact types and sets of geometry with contact relationships (this part is automatically identified and obtained by the program). Boundary contact types include endpoint-face (i.e., the end point of the rod contacts the solid surface), shell edge-face (i.e., the shell edge contacts the solid surface), surface-face (i.e., solid surfaces contact each other), and solid embedded (i.e., the component is embedded inside the solid). In step two, it is difficult to use the traditional manual modeling method, while using the geometric modeling interface to generate a whole multi-scale geometric model can solve the problems of cumbersome modeling and easy loss of component information.

[0076] In this invention, selecting appropriate constraint relationships at the interface between macroscopic and microscopic geometric components can ensure the true continuity of the component boundaries.

[0077] Step 3: Establish an analytical model

[0078] A global multi-scale finite element model is established using multi-scale geometric models, boundary contact geometric information, and data from the overall rod-shell computational model. This model includes second micro-units and second macro-units that correspond one-to-one with the first micro-units and first macro-units in the overall rod-shell computational model. The second micro-units contain the internal forces of the second member. The global multi-scale finite element model includes information such as period and inter-story displacement; this information is standard data in multi-scale finite element models and will not be described in detail here.

[0079] Specifically, a finite element modeling program is written using the Python scripting function in the commercial finite element analysis software ABAQUS. The multi-scale geometric model, boundary contact geometric information, and overall rod-shell calculation model data are output according to the Python script rules to obtain the ABAQUS script file of the overall multi-scale finite element model. Running the script in ABAQUS generates the overall multi-scale finite element analysis model. Using the Python scripting provided by ABAQUS to write the finite element modeling program can solve the problem of cumbersome pre-processing steps for finite element models.

[0080] Step 4: Model Validation

[0081] To determine the rationality of the overall multi-scale finite element analysis model, calculate the percentage difference between the period of the overall multi-scale finite element model and the period of the overall rod-shell calculation model, the percentage difference between the inter-story displacement of the overall multi-scale finite element model and the inter-story displacement of the overall rod-shell calculation model (these values ​​can be directly extracted from the calculation results of the overall multi-scale finite element model and the overall rod-shell calculation model), and the percentage difference between the internal forces of the second component and the internal forces of the first component. If the percentage differences of the period and the inter-story displacement are both less than 5%, and the percentage difference between the internal forces of the second component and the internal forces of the first component is less than 10%, then the overall multi-scale finite element model is considered reasonable, and the verification of the overall multi-scale finite element model is complete. Otherwise, repeat steps two to three to fine-tune the geometric components and boundary contact geometry of the micro-unit parts until the overall multi-scale finite element model meets the above conditions.

[0082] The percentage of the difference between periods is calculated using the following formula:

[0083] The percentage of the difference between periods = |T2 / T1-1|

[0084] Where T2 is the period of the overall multi-scale finite element model, and T1 is the period of the overall rod-shell calculation model.

[0085] The percentage of the difference in inter-story drift is calculated using the following formula:

[0086] The percentage of the difference in inter-story drift = |d2 / d1-1|

[0087] Where: d2 is the inter-story displacement of the overall multi-scale finite element model, and d1 is the inter-story displacement of the overall rod-shell calculation model.

[0088] The ratio of the difference between the internal forces of the second component and the internal forces of the first component is calculated using the following formula:

[0089] The percentage difference between the internal forces of the second component and the internal forces of the first component = |N2 / N1-1|

[0090] Where: N2 is the internal force of the second component of the second micro-unit in the overall multi-scale finite element model, and N1 is the internal force of the first component of the first micro-unit in the overall rod-shell calculation model.

[0091] In this invention, the information contained in the overall multi-scale finite element model and the overall rod-shell calculation model are corresponding; for example, the period and inter-story displacement in the two models are corresponding.

[0092] Step 5: Analyze and output the results

[0093] The rationality of the structural system, components and nodes is judged based on the reasonable overall multi-scale finite element model obtained in step four. Rationality includes whether the overall structural indicators (period and inter-story drift angle), component stiffness and strength, and the stress and strain levels of the nodes all meet (within the allowable range) the design objectives. For example, the "Code for Seismic Design of Buildings GB50011-2010" and the "Technical Specification for Concrete Structures of High-Rise Buildings JGJ3-2010" stipulate that different structural types need to meet different inter-story drift angle limits, and the structural period will affect the selection of the seismic influence coefficient. The allowable strengths of concrete and reinforcing steel in the "Code for Design of Concrete" (GB50010-2010) and the allowable strength values ​​of steel in the "Code for Design of Steel Structures" (GB50017-2017) both limit the allowable stress values ​​of finite element components. The "Code for Design of Steel Structures" (GB50017-2017) limits the stiffness of components through the slenderness ratio to prevent component instability. If the above conditions are met, the analysis ends and outputs the results; otherwise, steps two through four are repeated to fine-tune the geometry or nodes of the micro-units until the above rationality conditions are satisfied.

[0094] The fine-tuning in this invention includes determining whether the ends of macroscopic geometric components and the end faces of microscopic geometric components are actually aligned. If not, an alignment operation is performed, and then the element type and mesh of the geometric components are reselected, as well as the boundary contact geometry type is selected. If they are aligned, no fine-tuning is performed.

[0095] Taking a super high-rise project as an example, this paper illustrates the specific operation process and effects of the multi-scale finite element analysis method proposed in this invention. The selected analysis case consists of a 450-meter-high multi-functional integrated tower and a commercial podium. The tower has 100 floors above ground, with a Y-shaped core tube in the center. The tower's lateral force resisting structure is a hybrid frame-core tube-outrigger truss system. The three single towers above the 18th floor are connected by three huge connected trusses to form an overall anti-overturning effect. The floors where the connected trusses are located are L46-51, L76-79, and L97-100, respectively. Typical floor plans and elevation partitions of the structure are shown below. Figures 2-4 .

[0096] Taking this project as an example, under seismic loading, the overall bending deformation of the tower causes vertical displacement deformation of the mega-columns, which may lead to vertical and horizontal shear failure at the connection nodes of the chords and web members of the truss, as well as bending failure at the connection between the chords and the internal steel sections. Therefore, the connected trusses from floors L46 to L51 and the shear wall members of those floors are selected as micro-units, while the remaining components of the entire tower are selected as macro-units. In Rhino, the geometry generation program uses the file function to read the overall tower shell calculation model data and generate a three-dimensional overall tower shell geometric model (e.g., Figure 5Using the solid model function, select all components in the aforementioned micro-unit area, and then click "Generate Solid" to obtain the overall multi-scale geometric model (e.g., Figures 6 to 7 The geometric program detects and records the boundary contact geometry information of macroscopic and microscopic geometric components, including the contact between the end of the rod / column and the end face of the solid wall (e.g., Figure 8 ), the contact between the shell wall edge and the solid wall end face (e.g. Figure 9 ), contact between solid end faces (e.g.) Figure 10 ) and the built-in contact between the solid wall and some truss member nodes (such as Figure 11 ).

[0097] The finite element modeling program compiles the tower's overall multi-scale geometric model information and the overall skeleton calculation model data into a Python script, imports it into ABAQUS, and obtains the tower's overall multi-scale finite element analysis model (e.g., ...). Figures 12 to 13 By comparing the overall multi-scale finite element model and the overall skeleton calculation model of the tower, the period (as shown in Table 1, the difference in period percentage is less than 5% in all three period calculations), inter-story displacement (as shown in Table 2, which shows the percentage difference in inter-story displacement for each of the 100 stories, and as can be seen from the figure, all are less than 5%), and the percentage difference in internal forces of the components (as shown in Table 2) were calculated. Figures 14-1 to 14-3 As shown, where Figure 14-1 This demonstrates the internal forces of the second component in the second micro-element of the multi-scale finite element model. Figure 14-2 This demonstrates the internal forces of the first component in the first micro-unit of the integral rod-shell computational model. The internal forces at corresponding positions are calculated using the aforementioned ratio of differences in internal forces, yielding the ratio of these differences for each micro-unit. Figure 14-3 It can be seen that the differences are all less than 10%, with the former accounting for less than 5% and the latter less than 10%. Therefore, the overall multi-scale finite element model of the tower can be considered reasonable. The stress cloud diagram of the entire connected truss (e.g., Mises stress cloud diagram) shows that... Figures 15-1 to 15-5 As can be seen, under the action of a rare earthquake in the X direction, a large area of ​​the web at the junction of the chord and web members has yielded, with the maximum stress in the yield zone reaching 498.3 MPa, exceeding the yield strength of Q550 (490 MPa). The stress levels at other nodes are all lower than the yield strength of this material. This can be seen from the principal stress contour map of the shear wall (e.g., ...). Figure 16-1 and Figure 16-2The results show that under the action of a rare earthquake in the X direction, the maximum value of the principal tensile stress in the shear wall is less than the standard value of the tensile strength of C70, and the principal compressive stress is less than the standard value of the compressive strength of C70. The shear wall concrete did not yield. Analysis of the above results indicates that the connection nodes between the chord and web members need further strengthening. This can be achieved by changing the force transmission path of the web member flange or by increasing the thickness of the web plate in the node area, thereby improving the shear resistance of truss node one. Therefore, we return to Rhino to fine-tune the style of the chord and web member connection nodes in the micro-unit area. The fine-tuning style is as follows: Figures 17-1 to 17-6 As shown. The fine-tuned calculation results are as follows. Figures 18-1 to 18-6 As shown, supplementary nodes three, four, and five, by connecting the flanges of the two web members, can offset the axial forces of the tension and compression web members. Therefore, the supplementary nodes can improve the strength and stiffness of the entire truss joint. The overall multi-scale finite element analysis of the tower is now complete.

[0098] Table 1

[0099] Model Name First period (s) Second period (s) Third period (s) Integral rod shell calculation model 8.802 8.761 7.314 Overall multi-scale finite element model 8.459 8.448 7.018 Formula for calculating the percentage of difference |T2 / T1-1| |T2 / T1-1| |T2 / T1-1| Percentage of differences (%) 4 3.7 4.2

[0100] Table 2

[0101]

[0102]

[0103] The beneficial effects of this invention are:

[0104] (1) The simple constraints of the sub-model boundaries in the traditional finite element analysis method cause the boundary conditions to be distorted. By establishing an overall structural model, the real boundary conditions of the key analysis parts can be simulated, and a more realistic overall model stiffness matrix can be obtained.

[0105] (2) Multiple scale finite elements are flexibly adopted. Micro finite elements are used for key components and macro finite elements are used for non-key components. Therefore, it can simplify the calculation of the degrees of freedom of the overall model and reflect the spatial stress state of the key components.

[0106] (3) Compared with traditional finite element modeling methods, using existing structural model information reading and writing interfaces to model greatly improves modeling efficiency and automatically and quickly completes element material definition, multi-scale element group interaction constraints, load application and boundary condition application.

Claims

1. A method for analyzing building structures based on multi-scale finite element method, characterized in that: Includes the following steps: Step 1: Determine the calculation objectives, including establishing a calculation model of the overall rod and shell to be analyzed, and dividing the parts in the building structure into two types of analysis parts: micro-unit parts and macro-unit parts; Step 2: Establish a geometric model. Based on the geometric data in the overall rod and shell calculation model, perform three-dimensional modeling to obtain a three-dimensional overall rod and shell geometric model. Generate macroscopic and microscopic geometric components according to the type of analysis part, and process the interface between the two types of components to obtain a multi-scale geometric model of the whole. Step 3: Establish the analysis model. A multi-scale finite element model is established using multi-scale geometric models, boundary contact geometric information, and overall rod-shell calculation model data. Step 4: Model Verification. Determine the rationality of the overall multi-scale finite element analysis model. Calculate the period, inter-story displacement, and internal forces of the components in the micro-unit parts of the overall multi-scale finite element model and the overall rod and shell calculation model. If the difference between the period and inter-story displacement is less than 5%, and the difference between the internal forces of the components in the micro-unit parts is less than 10%, then the overall multi-scale finite element model is considered reasonable, and the verification of the overall multi-scale finite element model is completed. Otherwise, repeat steps 2 to 3 to fine-tune the geometric components and boundary contact geometry of the micro-unit parts until the overall multi-scale finite element model meets the above conditions. The percentage of the difference between periods is calculated using the following formula: The percentage of the difference between periods = |T2 / T1-1| Where: T2 is the period of the overall multi-scale finite element model, and T1 is the period of the overall rod-shell calculation model; The percentage of the difference in inter-story drift is calculated using the following formula: The percentage of the difference in inter-story drift = |d2 / d1-1| Where: d2 is the inter-story displacement of the overall multi-scale finite element model, and d1 is the inter-story displacement of the overall rod-shell calculation model; The proportion of the internal force difference in the micro-unit parts of the component is calculated using the following formula: The percentage difference between the internal forces of the second component and the internal forces of the first component = |N2 / N1-1| Where: N2 is the second component internal force in the micro-unit part of the overall multi-scale finite element model, and N1 is the first component internal force in the micro-unit part of the overall rod-shell calculation model; Step 5: Analyze and output the results. Based on the reasonable overall multi-scale finite element model obtained in Step 4, determine the rationality of the structural system, components, and nodes. Rationality includes whether the overall structural indicators, component stiffness and strength, and stress-strain levels of the node units all meet the design objectives. If they meet the requirements, end the analysis and output the results. Otherwise, repeat Steps 2 to 4 to fine-tune the geometric components or nodes of the micro-units until the rationality conditions are met.

2. The building structure analysis method based on multi-scale finite element method according to claim 1, characterized in that: Step one specifically involves establishing an overall rod-shell calculation model of the building structure to be analyzed. Based on the complexity of the components or nodes of the overall rod-shell calculation model and the nonlinearity of the component materials during loading, it is divided into two types of analysis parts: micro-unit parts and macro-unit parts.

3. The building structure analysis method based on multi-scale finite element method according to claim 2, characterized in that: In terms of node complexity, nodes with more than 4 converging components or node domain size greater than 1.5 times the component cross-sectional size, and in terms of nonlinearity, components and nodes whose material stress exceeds the yield strength or whose material strain exceeds the ultimate strain, are defined as micro-unit parts, and the remaining parts are macro-unit parts.

4. The building structure analysis method based on multi-scale finite element method according to claim 2, characterized in that: Step two includes: Three-dimensional modeling is performed based on the geometric data in the integral rod-shell calculation model to obtain the three-dimensional integral rod-shell geometric model. Based on the type of analysis area, macroscopic and microscopic geometric components are generated, and the interface between the two types of components is processed to obtain a comprehensive multi-scale geometric model. Boundary contact geometric information is established and stored based on the boundary contact conditions of the macroscopic and microscopic geometric components.

5. The building structure analysis method based on multi-scale finite element method according to claim 1, characterized in that: Step 3: Establish the analysis model by generating it using a Python script in the commercial finite element analysis software ABAQUS.

6. The building structure analysis method based on multi-scale finite element method according to claim 1, characterized in that: The fine-tuning includes determining whether the ends of macroscopic geometric components and the end faces of microscopic geometric components are actually aligned; otherwise, an alignment operation is performed, and then the element type and mesh of the geometric components are reselected, as well as the boundary contact geometry type is selected.

Images