A design method for support nodes of composite materials and dissimilar materials

Abstract

This invention discloses a design method for composite material and dissimilar material support joints: A continuous shell element model of the composite material and dissimilar material support joint is established; a web thickness matrix and a height matrix are established, as well as a matrix of web fiber volumes corresponding to different models; the vertical displacement degrees of freedom are coupled to each node in the loading area of ​​the steel connector's upper surface, and vertical loads are applied; the bottom of the support is set as a fixed constraint, and corresponding loads are applied to the model; the matrix of the ultimate bearing capacity of the model under the ultimate load condition is obtained; a surrogate model that can represent the input-output relationship is constructed; the ultimate bearing capacity of the model per unit volume is established as the objective function, and the maximum value of the ultimate bearing capacity of the model per unit volume that satisfies the conditions is searched; the web thickness and web height corresponding to this extreme value are the optimal structural design of the support joint. This invention improves the design efficiency of support joints, maximizes the utilization rate of composite materials, and makes the structural design more rational.

Description

Technical Field

[0001] This invention relates to a design method for support nodes made of composite materials and dissimilar materials, belonging to the field of ship structural design. Background Technology

[0002] Composite material and dissimilar material support structures are among the main load-bearing structures of the main hull. Currently, the design of support structures mainly considers strength. Based on finite element simulation, the composite material and dissimilar material support structures are optimized by changing the height of the cap profile web or the length of the panel. This method requires repeatedly building a large number of node models by changing the web height and web length, resulting in low optimization efficiency. At the same time, the model designed by this method does not consider the utilization rate of composite materials, and cannot meet the requirement of maximizing material utilization in structural design.

[0003] Composite material and dissimilar material support structures are among the main load-bearing structures of a ship's hull. Currently, the optimization design of these support structures mainly considers strength. Based on finite element simulation, the composite material and dissimilar material support structures are optimized by changing the web height or panel length of the cap profile. This method requires repeatedly establishing a large number of node models by changing parameters such as web height and web length, resulting in low optimization efficiency. Furthermore, the models designed by this method do not consider the utilization rate of composite materials and do not evaluate the relationship between the web thickness and web height and the strength of the support nodes from the perspective of the ultimate load-bearing capacity per unit volume of composite fiber. This fails to meet the requirement of maximizing material utilization in engineering structural design. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a design method for support nodes made of composite materials and dissimilar materials.

[0005] To address the above problems, this invention provides a method for designing support joints made of composite materials and dissimilar materials, comprising the following steps:

[0006] Step 1): Establish a continuous shell element model for the support joint of composite materials and dissimilar materials;

[0007] Step 2): Considering the main structural dimensions affecting node strength – the web thickness and web height of the cap profile – establish the web thickness matrix as follows: , can be represented as The corresponding web height matrix is , can be represented as The node model, in mm, corresponds to the model These can be represented by matrices as follows:

[0008] ;

[0009] Step 3): Assume the length of the support node is...l The web fiber volume corresponding to different models It can be represented as a matrix:

[0010] ;

[0011] Model in Matrix a i,j The corresponding volume v i,j It can be represented as v i,j = l × d i × h j ,in i = 1 , 2 … m , j = 1 , 2 … n ;

[0012] Step 4): Connect the steel connector to the contact surface of the composite panel, couple the vertical displacement degrees of freedom of each node in the loading area on the upper surface of the steel connector, apply vertical loads, set the bottom of the support column as a fixed support constraint, and apply the corresponding loads to the model.

[0013] Step 5): Obtain the model under ultimate load conditions. The corresponding ultimate bearing capacity It can be represented by a matrix as follows:

[0014] ;

[0015] Step 6): Model The corresponding web thickness and web height d i , h j As inputs, the web volume and the model's ultimate bearing capacity are... v i,j , f max i,j As output, a total of Using training samples, construct a proxy model that can represent the input-output relationship;

[0016] Step 7): Establish the ultimate bearing capacity of the model per unit volume under the objective function. The maximum value of the ultimate bearing capacity of the model per unit volume that meets the conditions is searched. The web thickness and web height corresponding to this extreme value are the optimal structural design of the support node.

[0017] Preferably, in step 1), the continuous shell element model of the composite material and dissimilar material support node is established using ABAQUS finite element software.

[0018] More preferably, the continuous shell element model of the composite material and dissimilar material support node in step 1) is composed of composite material, foam core material, adhesive layer and dissimilar material connection, wherein the adhesive layer adopts zero-thickness cohesive element.

[0019] Furthermore, the composite material adopts continuous shell unit SC8R, the foam core material adopts solid unit C3D8R, the adhesive layer adopts COH3D8 unit, and the dissimilar material is I-beam Q235.

[0020] Preferably, in step 4), the steel connector and the contact surface of the composite material panel are bonded together.

[0021] Preferably, in step 4), dynamic display analysis is used when applying the corresponding load to the model.

[0022] Preferably, in step 5), the model under the ultimate load condition is obtained based on ABAQUS finite element analysis software. The corresponding ultimate bearing capacity .

[0023] Preferably, in step 6), the relevant mapping relationship between input variables and output is first established, and then a proxy model that can represent the input-output relationship is constructed.

[0024] More preferably, in step 6), a machine learning algorithm is used to establish a correlation mapping relationship between input variables and output.

[0025] Preferably, in step 7), an artificial intelligence algorithm is used to search for the maximum value of the ultimate bearing capacity of the model per unit volume that meets the conditions.

[0026] This invention, with a fixed web length for the cap profile of the support node, uses the web height and thickness of the cap profile as input variables to be optimized, and the web volume and ultimate bearing capacity of the support node as output variables. It establishes a proxy relationship model between the web height and thickness of the cap profile and the web volume and ultimate bearing capacity of the support node. Using an artificial fish swarm algorithm with the ultimate bearing capacity of fibers per unit volume as the objective function, it quickly obtains the optimal web thickness and height that satisfy the extreme value of the objective function. This method can improve the design efficiency of support nodes, meet the requirement of maximizing material utilization in structural design, and achieve optimal design of composite material and dissimilar material support structures.

[0027] This invention uses the web height and web thickness of the cap profile of the support node as input variables to be optimized, and the web volume and ultimate bearing capacity of the support node as output variables. Based on machine learning algorithms, a proxy relationship model is established between the web height and web thickness of the cap profile of the support node and the web volume and ultimate bearing capacity of the support node. Then, with the help of intelligent algorithms, the optimal web thickness and web height that satisfy the extreme value of the objective function are quickly obtained using the ultimate bearing capacity of fiber per unit volume as the objective function. This improves the design efficiency of support nodes, avoids repetitive modeling work, meets the requirement of maximizing material utilization in structural design, and realizes the optimal design of composite material and dissimilar material support structures.

[0028] This invention improves the design efficiency of support joints, maximizes the utilization rate of composite materials, and makes the structural design more rational. Compared with the prior art, the beneficial effects of this invention are:

[0029] (1) Establish a proxy model of web height and thickness and web fiber volume and unit volume ultimate bearing capacity. The optimal web thickness and web height are obtained quickly through intelligent algorithm, which improves the design efficiency of support node and avoids repetitive modeling work.

[0030] (2) An evaluation index for the ultimate bearing capacity per unit fiber volume is proposed, which not only meets the strength requirements, but also improves the material utilization rate in the design of composite material structures, making the structural design more reasonable. Attached Figure Description

[0031] Figure 1 A flowchart illustrating the design method for composite material and dissimilar material support nodes provided by this invention;

[0032] Figure 2 The support node for connecting composite materials and dissimilar materials was established for the example.

[0033] Figure 3 The mesh model established in step (2) of the embodiment;

[0034] Figure 4 This is a schematic diagram of the model dimensions involved in step (3) of the embodiment;

[0035] Figure 5 This is a structural diagram of dissimilar materials (I-beams);

[0036] Figure 6 This is a schematic diagram of the composite material structure;

[0037] Figure 7 This is a schematic diagram of the structure of the foam core material. Detailed Implementation

[0038] To make the present invention more apparent and understandable, preferred embodiments are described in detail below with reference to the accompanying drawings.

[0039] Example

[0040] A design method for support nodes of composite materials and dissimilar materials:

[0041] (1) Establish such Figure 1 The support node shown is a connection between a composite material and a dissimilar material. The upper part of the model is an I-beam made of Q235 steel, the composite material is glass fiber, the core material is PVC foam sandwich, and the adhesive layer at the connection uses zero-thickness cohesive elements.

[0042] (2) Steel connectors use eight-node hexahedral elements C3D8R, composite materials use eight-node quadrilateral in-plane universal continuous shell elements SC8R, foam core materials use eight-node quadrilateral in-plane universal continuous shell elements SC8R, and adhesive layers use eight-node three-dimensional bonding elements COH3D8, establishing a system as follows: Figure 2 The mesh model shown (which is composed of, for example) Figure 6 The composite material shown, such as Figure 7 The foam core material and adhesive layer shown are as follows: Figure 5 The dissimilar materials shown are connected according to design requirements.

[0043] (3) Model dimensions are as follows Figure 3 As shown, the web thickness matrix is ​​established as follows: , represented as The corresponding web height moment is , can be represented as The node model. This embodiment uses 200 sample points, of which the number of web thickness variables is... m The web height variable is 20. h The thickness is 10. d m The value range is 10~30, and the height is... h n The value ranges from 80 to 150, and the unit is mm. The corresponding model... These can be represented by matrices as follows:

[0044] ;

[0045] Establish the web length of the support node as a fixed value. l The web fiber volume corresponding to different models It can be represented as a matrix:

[0046] ;

[0047] Model in Matrix a i,jThe corresponding volume v i,j It can be represented as v i,j = l × d i × h j ,in i = 1 , 2 … m , j = 1 , 2 … n .

[0048] (4) The bottom of the support node is coupled to the reference point RP1. A fully fixed constraint is applied to the reference point RP1 as shown in the figure. The 800mm×300mm range in the middle of the top plate of the steel connector is coupled to the reference point RP2. A vertical load is applied to the reference point RP2. A smooth analysis step is used. The ultimate condition of the support model is calculated by using the Visual Studio associated subroutine Fortran to obtain the failure condition of the support node under the ultimate condition.

[0049] (5) Based on ABAQUS finite element analysis software, obtain the model under ultimate load conditions. The corresponding ultimate bearing capacity It can be represented by a matrix as follows:

[0050] ;

[0051] (6) Model The corresponding web thickness and web height ( d i , h j As input, the web volume and the model's ultimate bearing capacity ( v i,j , f max i,j As output, there are a total of 200 training samples, each containing four data points ( ). d i , h j , v i,j , f max i,j A multi-kernel Gaussian process regression model was used to establish the relationship between web size parameters and volume and ultimate bearing capacity for different samples.

[0052] (7) Establish the ultimate bearing capacity of the model per unit volume under the objective function The Artificial Fish Swarm Algorithm (ASFA) was used to find the maximum value of the ultimate bearing capacity of the model per unit volume. The parameters were set as follows: field of view size... Visual = 8, Crowding factor Search step size S = 10, number of foraging attempts Try_N = 25, number of fish N = 30. The final set of model parameters obtained is (23.68mm, 102.64mm). At this point, the model has the maximum unit fiber volumetric load-bearing capacity. The web thickness of 23.68mm and the web height of 102.64mm are the optimal structural design dimensions for this support node model. The basic process is as follows: Figure 4 As shown.

Claims

1. A method for designing support joints of composite materials and dissimilar materials, characterized in that, Includes the following steps: Step 1): Establish a continuous shell element model for the support joint of composite materials and dissimilar materials; Step 2): Considering the main structural dimensions affecting node strength – the web thickness and web height of the cap profile – establish the web thickness matrix as follows: , can be represented as The corresponding web height matrix is , can be represented as The node model, in mm, corresponds to the model These can be represented by matrices as follows: ； Step 3): Assume the length of the support node is... l The web fiber volume corresponding to different models It can be represented as a matrix: ； Model in Matrix a i,j The corresponding volume v i,j It can be represented as v i,j = l × d i × h j ,in i = 1 , 2 … m , j = 1 , 2 … n ; Step 4): Connect the steel connector to the contact surface of the composite panel, couple the vertical displacement degrees of freedom of each node in the loading area on the upper surface of the steel connector, apply vertical loads, set the bottom of the support column as a fixed support constraint, and apply the corresponding loads to the model. Step 5): Obtain the model under ultimate load conditions. The corresponding ultimate bearing capacity It can be represented by a matrix as follows: ； Step 6): Model The corresponding web thickness and web height d i , h j As inputs, the web volume and the model's ultimate bearing capacity are... v i,j , f max i,j As output, a total of Using training samples, construct a proxy model that can represent the input-output relationship; Step 7): Establish the ultimate bearing capacity of the model per unit volume under the objective function. The maximum value of the ultimate bearing capacity of the model per unit volume that meets the conditions is searched. The web thickness and web height corresponding to the maximum value are the optimal structural design of the support node.

2. The design method for composite material and dissimilar material support joints as described in claim 1, characterized in that, In step 1), the ABAQUS finite element software is used to establish a continuous shell element model of the composite material and dissimilar material support node.

3. The design method for composite material and dissimilar material support joints as described in claim 2, characterized in that, The continuous shell element model of the composite material and dissimilar material support node in step 1) is composed of composite material, foam core material, adhesive layer, and dissimilar material connection, wherein the adhesive layer adopts zero-thickness cohesive element.

4. The design method for composite material and dissimilar material support joints as described in claim 3, characterized in that, The composite material uses continuous shell units SC8R, the foam core uses solid units C3D8R, the adhesive layer uses COH3D8 units, and the dissimilar material is Q235 I-beams.

5. The design method for composite material and dissimilar material support joints as described in claim 1, characterized in that, In step 4), the steel connector and the composite material panel are bonded together.

6. The design method for composite material and dissimilar material support joints as described in claim 1, characterized in that, In step 4), dynamic display analysis is used when applying the corresponding load to the model.

7. The design method for composite material and dissimilar material support joints as described in claim 1, characterized in that, In step 5), the model under the ultimate load condition is obtained based on the ABAQUS finite element analysis software. The corresponding ultimate bearing capacity .

8. The design method for composite material and dissimilar material support joints as described in claim 1, characterized in that, In step 6), the relevant mapping relationship between input variables and output is first established, and then a proxy model that can represent the input-output relationship is constructed.

9. The design method for composite material and dissimilar material support joints as described in claim 8, characterized in that, In step 6), a machine learning algorithm is used to establish a relevant mapping relationship between input variables and output.

10. The design method for composite material and dissimilar material support joints as described in claim 1, characterized in that, In step 7), an artificial intelligence algorithm is used to search for the maximum value of the ultimate bearing capacity of the model per unit volume that meets the conditions.

Images