CN112199876A

CN112199876A - Energy absorption structure optimization method with expected force response course as target

Info

Publication number: CN112199876A
Application number: CN202011114159.8A
Authority: CN
Inventors: 安伟刚; 王世根; 韩煦
Original assignee: Northwestern Polytechnical University
Current assignee: Northwestern Polytechnical University
Priority date: 2020-10-18
Filing date: 2020-10-18
Publication date: 2021-01-08
Anticipated expiration: 2040-10-18
Also published as: CN112199876B

Abstract

An energy absorption structure optimization method taking expected force response history as a target is characterized in that a proper expected force response curve is set, the value of an objective function in the optimization process is continuously updated in an iteration mode to obtain a new finite element model, the actual force response curve of the structure can be gradually close to the expected force response curve, and the energy absorption capacity of the structure is improved. The invention not only can change the deformation mode of the thin-wall pipe from Euler deformation with low energy absorption to progressive buckling deformation with high energy absorption, but also can simultaneously improve a plurality of energy absorption evaluation indexes. The actual force response curve of the optimized thin-walled tube is closer to the expected force response curve. The energy absorption part obtained by the invention can generate an ideal deformation mode, dissipates more impact kinetic energy and has great guiding significance for the design of energy absorption devices of various vehicles.

Description

Energy absorption structure optimization method with expected force response course as target

Technical Field

The invention belongs to the technical field of traffic safety, and particularly relates to an energy absorption structure optimization method taking expected force response course as a target.

Background

Safety is the most fundamental requirement in the vehicle design process. In recent years, with the development of transportation industry, collision accidents have become a prominent problem affecting the life safety of passengers. Thin-walled metal structures have found widespread use in energy absorbing structures in the automotive and aerospace industries due to their light weight, manufacturability, and good impact resistance. When a vehicle or an airplane collides, the energy absorption structure absorbs impact energy through plastic deformation to ensure the safety of passengers to the maximum extent. The energy absorption capacity of the thin-wall metal structure is closely related to the deformation mode of the thin-wall metal structure, the ideal deformation mode is progressive buckling, and the mode can generate more fold deformation to dissipate impact kinetic energy.

Currently, researchers have made a lot of research on energy absorbing structures, and most of the research is directed to the optimization design of element-level dimensions. The size of the element level is optimized, the number of design areas and design variables is small, and the optimization effect is inevitably influenced. The free dimension optimization method is a generalized dimension optimization method which takes the unit thickness in a thin-wall structure finite element model as a design variable and solves the optimal thickness distribution, and can conveniently carry out the unit-level optimization design of the whole structure system. However, the method has huge design variables, usually from tens of thousands to hundreds of thousands, and the impact dynamics problem has strong nonlinearity, and the single display dynamics analysis takes long time. In order to solve the above problems, the following two methods have been developed. The method effectively reduces the calculated amount, enables the traditional Optimization method to be adopted, but a plurality of static loads are difficult to well equivalent actual impact load working conditions, the accuracy of a dynamic analysis result is poor, and the Optimization result is influenced. Secondly, the hybrid cellular automata method for designing the structure crashworthiness is published by Andre s Tovar in the article of "addressing for crashworthiness addressing using hybrid cellular automata" in International Journal of Vehicle Design, and the method adopts display dynamics analysis, so that the analysis result is accurate, but the adopted "energy distribution criterion" is unreasonable, the "regional energy distribution criterion" excessively depends on the prior experience, and the optimization result has uncertainty.

Disclosure of Invention

In order to overcome the defect that an optimization result in the prior art has uncertainty, the invention provides an energy absorption structure optimization method taking expected force response course as a target.

The specific process of the invention is as follows:

step 1, establishing an original finite element model of the thin-wall pipe under impact:

and according to the structural characteristics of the impacted thin-wall pipe and the rigid impact plate, obtaining a geometric model of the thin-wall pipe and the rigid impact plate, and establishing an original finite element model of the thin-wall pipe under impact.

The structure characteristics of the impacted thin-wall pipe and the rigid impact plate are divided into two parts, wherein one part is the thin-wall pipe with a square cross section, and the other part is the rectangular rigid impact plate; the rigid impact plate is positioned at one end face of the thin-walled tube, and the central line of the thin-walled tube in the length direction is superposed with the geometric center of the rigid impact plate, or the geometric central line of the cross section of the thin-walled tube is superposed with the central line of the rigid impact plate in the width direction; the minimum distance between the end face of the thin-walled tube and the surface of the rigid impact plate is 5 mm.

When the central line of the thin-wall pipe in the length direction is coincident with the geometric center of the rigid impact plate, the impacted thin-wall pipe and the rigid impact plate are equally divided into four parts along the length direction, and one part is taken to obtain 1/4 thin-wall pipe and rigid impact plate structures which are used as geometric models for establishing the original finite element models of the impacted thin-wall pipes.

When the geometric center line of the cross section of the thin-wall pipe is coincident with the center line of the rigid impact plate in the width direction, the impacted thin-wall pipe and the rigid impact plate are divided into two parts along the length direction, and one part is taken to obtain 1/2 thin-wall pipe and rigid impact plate structures which are used as a geometric model for establishing an original finite element model of the impacted thin-wall pipe.

Dividing the thin-wall pipe in the obtained 1/4 geometric model or the thin-wall pipe in the 1/2 geometric model into a plurality of full-integral shell units through finite element software Hypermesh; dividing the rigid impact plate in the geometric model of 1/4 or 1/2 into several solid units.

Step 2, solving an actual force response curve of the thin-walled tube under impact:

and (3) performing simulated impact on the original finite element model established in the step (1) by utilizing display dynamics analysis software LS-DYNA, and obtaining an impact process of the rigid impact plate and a result file of deformation of the thin-walled tube, wherein the result file is contained in the original finite element model. And extracting impact force between the thin-walled tube and the rigid impact plate and displacement data of the rigid impact plate in the result file by using MATLAB software to obtain an actual force response curve when the thin-walled tube is impacted.

The actual force response curve is a relationship between impact force and displacement; the impact force is the impact force between the thin-walled tube and the rigid impact plate, and the displacement is the displacement of the rigid impact plate.

Step 3, setting an expected force response curve:

setting an expected force response curve by obtaining the peak value and the average value of the actual force response curve when the thin-walled pipe is impacted;

the expected force response curve is the relationship between impact force and displacement set in the optimization process; the impact force is the impact force between the thin-walled tube and the rigid impact plate, and the displacement is the displacement of the rigid impact plate.

The expected force response curve is set to be a straight line parallel to the x-axis, and the average value of the impact force of the expected force response curve is 5.00 multiplied by 10⁴N; or a broken line divided into two parts, one part is a straight line with a slope, the other part is a straight line parallel to the x-axis, and the slope of the straight line with the slope is 9.50 multiplied by 10 from the coordinate origin⁵The maximum expected force value is 7.00X 10⁴N; the expected force value of a straight line parallel to the x-axis is 7.00X 10⁴N; or a power function curve in which the expected force response increases with increasing displacement, corresponding to a maximum response force of 7.50 x 10 at 0.7m⁴And N is added. The above-mentionedIn a coordinate system of the expected force response curve, an abscissa x is the displacement of the rigid impact plate, and the unit is m; the ordinate F is the collision force between the thin walled tube and the rigid impingement plate in units of N.

Step 4, determining an objective function of the thin-wall pipe optimization process:

the objective function is the minimum of the difference between the actual force response curve and the desired force response curve, and is expressed as:

wherein minC is an objective function, F (x) is an actual response curve, F⁰(x) Is a target response force curve, x₀And x_eRespectively the initial displacement and the final displacement of the rigid striking plate during the impact.

Step 5, determining design variables:

the design variable is the thickness of each shell element during optimization of the energy absorbing structure. The optimization process is realized in an iterative mode; the thickness of each shell element varies in each iteration of the design.

The thickness of each shell unit of the thin-walled tube in the original finite element model is respectively used as the current design variable, namely the number of the design variables is the same as that of the shell units. The initial thickness of each shell element was 3 mm.

Step 6, determining constraint conditions:

the thickness of the shell element and the mass of the thin-walled tube in the original finite element model in the optimization process are taken as constraint conditions,

wherein t is a unit thickness matrix of the finite element model thin-walled tube; t is t_maxAnd t_minUpper and lower limits for the thickness of the shell element, respectively; m is the mass of the 1/4 thin-wall pipe in the optimization process; m is₀1/4 the mass of a thin-walled tube with a thickness of 3 mm.

An upper limit t of the thickness of the shell element_max5mm, lower limit t of the thickness of the shell element_min＝0.5mm。

Step 7, obtaining a new finite element model:

obtaining a new finite element model by obtaining the value of the objective function of the thin-walled tube optimization process in the step 4, specifically:

and 4, calculating the objective function value of the original finite element model by using the formula 4-1 in the step 4.

Establishing a relation F (x) between the impact force between the thin-wall pipe and the rigid impact plate in the original finite element model in the step 2 and the unit thickness,

wherein, Delta E_kFor the change in kinetic energy during impact, Δ E_IFor the purpose of the change of energy during the impact,

is the thickness of the nth cell in the ith pass,

is the thickness of the nth cell in the (i-1) th time, and N is the number of the cells; x is the initial displacement x of the rigid impact plate during impact₀Final displacement x from rigid impact plate_eAny position in between.

And (3) solving a formula 7-1 by using a constrained nonlinear multivariable optimization algorithm provided by MATLAB software to obtain the new thickness of each shell unit in the thin-wall tube, and replacing the thickness of each shell unit in the original finite element model established in the step 1 by using the new thickness to obtain a new finite element model.

Step 8, calculating an objective function value of the optimization process:

and obtaining a new objective function value again through the new finite element model.

And obtaining the new objective function value again in an iterative mode. The initial value of the number of iterations is 11 due to the presence of numerical noise during the kinetic analysis. The method comprises the following steps:

and returning to the step 2, and obtaining the actual force response curve of the new finite element model again by using the method in the step 2.

And step 7, calculating a new objective function value through the actual force response curve. In calculating the new objective function value, the desired force response curve, the expression of the objective function, the design variables and the constraint conditions are the same as the conditions and parameters determined in steps 3 to 6. And obtaining a new objective function value again, and obtaining a new finite element model again at the same time.

And continuously obtaining the actual force response curve of the new finite element model by using the new finite element model obtained again in the iteration.

And returning to the step 2 to obtain an actual force response curve of the new finite element model, entering the step 7, repeating the iteration mode for 10 times, and sequentially obtaining 10 new objective function values, namely obtaining one objective function value every time iteration is performed. Obtaining 11 objective function values in total, and numbering the objective function values as i in sequence; i is 1,2, … … 11.

I determining the minimum of 11 values of the objective function

C_i＝min[C₁,C₂,…,C₁₁]，

C_iIs the minimum of the 11 objective function values; i is the number of the minimum. If i is 1, the iteration stops; if i is more than 1, continuing to iterate for i-1 time; in the process of iterating for i-1 times, obtaining i-1 objective function values; the total number of objective function values is 11+ i-1.

II, taking the minimum value C of the ith, i +1, … i +10 in the 11+ i-1 objective function values_jJ is the number of the minimum value; the method comprises the following steps:

C_j＝min[C_i,C_i+1,…,C_i+10],j≥i，

if j ═ i, the iteration stops;

if j is larger than i, continuing to iterate j-i times, and obtaining j-i objective function values in the process of iterating j-i times; and obtaining 11+ i-1+ j-i objective function values in total.

III, repeating the process of obtaining the j-i objective function value in the II until the minimum value of the obtained current 11 objective functions is equal to the first value of the current 11 objective functions, and stopping iteration; if the iteration number reaches 50 times, the minimum value in the current 11 objective functions still cannot be equal to the first value in the current 11 objective functions, and the iteration is also stopped. So far, the optimization of the energy absorption structure is completed.

The peak force of the optimized energy absorption structure is 7.37-8.81 multiplied by 10⁴N, average force of 4.71-6.32 x 10⁴N, the collision load efficiency is 61.9-72.6%, and the specific energy absorption is 8.4-10.8 kJ/kg.

Compared with the prior art, the invention has the following beneficial effects:

when evaluating the energy absorption capacity of the energy absorption structure, four core evaluation indexes are available: peak force, average force, impact load efficiency and specific energy absorption. The peak force is the maximum force generated during the entire impact process, and in order to avoid serious injury to the occupant, the lower the peak force, the better. The average force is the average of the forces generated throughout the impact process, with the higher the average, the more energy absorbed. The crash load efficiency is the ratio of the average force to the peak force during the entire impact, ideally 100%, i.e. the force generated during impact remains constant. The specific energy absorption is the energy absorbed by the unit mass of the structure in the whole impact process, and the higher the specific energy absorption is, the stronger the energy absorption capability of the structure is. With the attached drawings, the beneficial effects obtained by the invention are as follows:

1. when one end of a thin-walled tube is fixed and the other end of the thin-walled tube is subjected to axial impact of a rigid impact plate, as shown in fig. 1, fig. 7 shows a force response curve 2 before optimization under the axial impact, the initial peak force is higher, and then the force response starts to be reduced, but the whole process still maintains higher average force. As shown in fig. 5, the thin walled tube undergoes a high energy absorbing progressive buckling deformation. As shown in Table 2, the peak force of the thin walled tube before optimization was 7.40X 10 throughout the impact process⁴N, average force 3.33X 10⁴N, the collision load efficiency is only 45.0 percent, and the specific energy absorption is 9.6 kJ/kg. To further improveThe energy absorption capacity of the thin-walled tube is set to be 5.00 multiplied by 10 as a constant value shown in figure 3⁴The curve of N is the expected force response curve, namely the average force of the thin-walled tube after the optimization is expected to reach 5.00 multiplied by 10⁴And N is added. The optimized actual force response curve, i.e., the optimized force response curve 3 under axial impact in FIG. 7, is closer to the desired force response curve, fluctuating up and down about it. The optimized thin walled tube also produces a high energy absorption progressive buckling deformation as shown in fig. 6. The values of the evaluation index of the thin-walled tube after optimization are given in Table 2, and the peak force is 7.37X 10⁴N, average force 4.71X 10⁴N, the collision load efficiency is 64.0 percent, and the specific energy absorption is 10.8 kJ/kg. Peak force reduction by 3.00 x 10²N, the average force is improved by 41.4%, the collision load efficiency is improved by 19 percentage points, and the energy absorption ratio is improved by 12.5%. The optimized thin-walled tube generates two thinner areas similar to a trigger device at the impacted end, and the characteristic is that the impacted end generates local buckling, so that the peak force is reduced. By setting a proper expected force response curve, the four evaluation indexes are simultaneously improved. As shown in fig. 4, the optimization process only performs 23 iterations, and obtains 23 objective function values 10 in the optimization process, and finds the minimum value 11 of the objective function in the optimization process at the 13 th iteration, so that the optimization efficiency is high.

2. As shown in fig. 8, when one end of the thin-walled tube is fixed and the other end of the thin-walled tube is subjected to an oblique impact of 15 degrees by the rigid impact plate, fig. 16 shows an actual force response curve 5 before optimization when the expected force response is a broken line, and as can be seen from fig. 13, the rigid plate moves forward by 0.1m at the beginning stage of the impact, 2 fold deformations are generated at the impacted end of the thin-walled tube, the thin-walled tube has a high impact force, and after 0.1m, the thin-walled tube has euler deformation, the energy absorption effect is reduced, and the impact force is sharply reduced to 1.00 × 10⁴N or less until the end of the impact. As shown in Table 3, the peak force of the thin walled tube during the entire impact was 7.96X 10⁴N, average force 1.66X 10⁴N, the collision load efficiency is only 20.8 percent, and the specific energy absorption is only 2.4 kJ/kg. In order to change the deformation mode of the thin walled tube, a desired force response curve is set as shown in fig. 11, the curve having a straight line with a slope at the front section and a straight line with a constant value at the rear section. Optimized actual force responseIn the case of the curve, i.e., the expected force response is a broken line in fig. 16, the optimized actual force response curve 6 is more closely fitted to the portion with a slope of the expected force response curve, and is constant at 7.00 × 10 in the expected force response curve⁴The part near the N fluctuates up and down, and after the rigid impact plate moves for 0.1m, the higher impact force level is still kept, so that the average force in the impact process is improved, and the energy absorption capacity of the structure is improved. The deformation mode of the optimized thin-walled tube is changed, as shown in fig. 14, 8 fold deformations are generated, and the euler deformation of low energy absorption is changed into the progressive buckling deformation of high energy absorption. The values of the evaluation index of the thin-walled tube after optimization are given in Table 3, and the peak force is 8.71X 10⁴N, average force of 6.32X 10⁴N, the collision load efficiency is 72.6 percent, and the specific energy absorption is 8.6 kJ/kg. The peak force is increased by 9.4%, the average force is improved by 280.7%, the collision load efficiency is improved by 52 percentage points, and the ratio of energy absorption is improved by 258.3%. Under the condition of increasing a small amount of peak force, the average force and the specific energy absorption are both improved by more than 250%. As shown in fig. 12, the optimization process only performs 22 iterations, and obtains 22 objective function values 10 in the optimization process, and finds the minimum value 11 of the objective function in the optimization process at the 12 th iteration, so that the optimization efficiency is high.

3. Under the impact condition of the thin walled tube in 2 above, the value of the objective function in the optimization process appears abrupt in the 6 th iteration as shown in fig. 12, resulting in the euler deformation with low energy absorption as shown in fig. 15. To avoid this during the optimization, a power function curve as shown at 17 is set to the desired force response curve, which increases with increasing displacement, corresponding to a maximum response force of 7.50 x 10 at 0.7m⁴And N is added. The optimized actual force response curve, that is, the optimized actual force response curve 9 when the expected force response is a power function curve under oblique impact in fig. 20, is very close to the expected force response curve, and fluctuates in a small range near the expected force response curve. After the rigid impact plate moves for 0.1m, the force response does not suddenly drop, and the expected force response curve is gradually increased, so that the energy absorption capacity of the structure is ensured. With reference to FIG. 19, the optimized thin walled tube produces a progression of high energy absorption during impact8 folds are formed by buckling to deform, and the energy absorption effect is more excellent. The values of the evaluation indices of the thin-walled tube after optimization are given in Table 4, with a peak force of 8.81X 10⁴N, average force of 5.45X 10⁴N, the collision load efficiency is 61.9 percent, and the specific energy absorption is 8.4 kJ/kg. The peak force is increased by 7.8%, the average force is improved by 228.3%, the collision load efficiency is improved by 41%, and the ratio of energy absorption is improved by 250.0%. By changing the power of the power function curve to lower the maximum value of the desired force response curve, the peak force can be reduced. As shown in fig. 18, the optimization process has performed 37 iterations to obtain 37 values 10 of the objective function in the optimization process, and the minimum value 11 of the objective function in the optimization process is found in the 27 th iteration, so that the objective function value does not suddenly change in the optimization process, and the optimization process is more stable.

4. The impact problem is a transient dynamics problem because the impact problem has strong nonlinearity. The solution of the problem needs to be carried out by display dynamics analysis software LS-DYNA, and the solution process is long in time. In the optimization process, each iteration needs to be performed once to analyze the display dynamics, and the optimization process only needs less iteration times, less than 50 times, so that the calculation time is saved.

In conclusion, by setting a proper expected force response curve, the deformation mode of the thin-wall pipe can be changed from Euler deformation with low energy absorption to progressive buckling deformation with high energy absorption, and a plurality of energy absorption evaluation indexes can be improved simultaneously. The actual force response curve of the optimized thin-walled tube is closer to the expected force response curve.

Therefore, the method can enable the actual force response curve of the structure to be gradually close to the expected force response curve in the optimization process, and improve the energy absorption capacity of the structure. The method overcomes the defects, the designed energy absorption part can generate an ideal deformation mode to dissipate more impact kinetic energy, and the method has great guiding significance for the design of energy absorption devices of various vehicles.

Drawings

FIG. 1 is a view of an axial impact model of example 1;

FIG. 2 is a diagram of a finite element model of embodiment 1;

FIG. 3 is the expected force response curve of example 1;

FIG. 4 is a graph of historical iterations of the objective function of example 1;

FIG. 5 is a deformation diagram before optimization of embodiment 1;

FIG. 6 is a modified diagram of the embodiment 1 after optimization;

FIG. 7 is a graph of the actual force response before and after optimization for example 1;

FIG. 8 is a schematic view of a model of oblique impact in examples 2 and 3;

FIG. 9 is a schematic view of a model of oblique impact in examples 2 and 3;

FIG. 10 is a diagram of a finite element model of examples 2 and 3;

FIG. 11 is the expected force response curve of example 2;

FIG. 12 is a historical iteration curve of the objective function of example 2;

FIG. 13 is a deformation diagram before optimization of examples 2 and 3;

FIG. 14 is a modified diagram of the embodiment 2 after optimization;

FIG. 15 is a deformation diagram in the optimization process of example 2;

FIG. 16 is a graph of the actual force response before and after optimization for example 2;

FIG. 17 is the expected force response curve of example 3;

FIG. 18 is a historical iteration curve of the objective function of example 3;

FIG. 19 is a modified diagram of the embodiment 3 after optimization;

FIG. 20 is a graph of the actual force response before and after optimization for example 3;

FIG. 21 is a flow chart of the present invention.

In the figure: 1. a desired force response straight line under axial impact; 2. force response curve before optimization under axial impact; 3. an optimized force response curve under axial impact; 4. a desired force response polyline under oblique impact; 5. under oblique impact, when the expected force response is a broken line, optimizing a force response curve before optimization; 6. under oblique impact, when the expected force response is a broken line, an optimized force response curve is obtained; 7. an expected force response power function curve under oblique impact; 8. under oblique impact, when the expected force response is a power function curve, optimizing the force response curve before optimization; 9. an optimized force response curve under oblique impact when the expected force response is a power function curve; 10. the value of the objective function in the optimization process; 11. the minimum value of the objective function in the optimization process.

Detailed Description

Example 1

The embodiment is an energy absorption structure optimization method taking a desired force response course as a target, and the specific process is as follows:

and establishing an original finite element model of the impacted thin-wall pipe according to the structural characteristics of the impacted thin-wall pipe and the rigid impact plate. As shown in fig. 1, the structural characteristics of the impacted thin-walled tube and the rigid impact plate are divided into two parts, one part is a thin-walled tube with a square cross section, the side length of the thin-walled tube is 100mm, the length of the thin-walled tube is 1000mm, and the wall thickness of the thin-walled tube is 3.0 mm; the other part is a square rigid impact plate with a side length of 200 mm. The rigid impact plate is positioned at one end face of the thin-walled tube, and the central line of the thin-walled tube in the length direction is superposed with the geometric center of the rigid impact plate; the end face of the thin-walled tube is 5mm away from the surface of the rigid impact plate. The end of the thin-wall pipe adjacent to the rigid impact plate is the front end.

The front end of the thin-walled tube is simulated to be impacted by the rigid impact plate. The impacted thin-wall pipe and the rigid impact plate are equally divided into four same parts in a shape like a Chinese character 'tian' along the length direction, the impact force born by each cross section of the thin-wall pipe is uniform when the thin-wall pipe is impacted, and one part of the thin-wall pipe and the 1/4 rigid impact plate are taken to obtain 1/4 thin-wall pipes and 1/4 rigid impact plate structures which are used as geometric models for building original finite element models of the thin-wall pipes under the impact. The thin-walled tube of 1/4 was divided into several shell elements by finite element software Hypermesh. To avoid hourglass deformation, all of the shell elements are fully integral shell elements. The rear end of the 1/4 thin-walled tube is fixed, so that the 1/4 rigid impact plate moves towards the 1/4 thin-walled tube at a constant speed of 5m/s, and the moving distance is 700 mm.

In this embodiment, each of the shell units has a size of 10 × 10mm and a wall thickness of 3.0mm, which is 1000 in total. The 1/4 rigid impact plates are solid units, and the number of the rigid impact plates is 4.

The properties of the thin-walled tube material are given in table 1 below.

Table 1: performance parameters of materials for thin walled tubes

Performance of	Density kg/m³	Modulus of elasticity Gpa	Poisson ratio	Yield strength Mpa
					Numerical value	7800	207	0.29	253

The force response curve determined in this example is a force-displacement curve.

And (3) performing simulated impact on the original finite element model established in the step (1) by using display dynamics analysis software LS-DYNA, and obtaining a result file containing the impact process of the 1/4 rigid impact plate and the deformation of the 1/4 thin-walled tube by using the display dynamics analysis software LS-DYNA. And (3) extracting impact force between the 1/4 thin-wall pipe and the 1/4 rigid impact plate and displacement data of the rigid impact plate in a result file by using MATLAB software to obtain an actual force response curve of the thin-wall pipe under impact.

Step 3, setting an expected force response curve:

and obtaining the peak value and the average value of the actual force response curve of the thin-wall pipe under the impact through the MATLAB software to obtain the expected force response curve.

In this embodiment, the peak value of the impact force of the actual force response curve is 7.40 × 10⁴N, average value 3.33X 10⁴And N is added. In the impact process, the average value of the impact force between the thin-wall pipe and the rigid impact plate reflects the energy absorption capacity of the thin-wall pipe. The higher the average value, the stronger the energy absorption capacity of the thin-walled tube. At the same time, the lower the peak force of the impact process, the better in order to avoid serious injury to the occupant.

In order to improve the energy absorption capacity of the thin-walled tube while preventing the occupant from being injured, the expected force response curve is arranged as a straight line parallel to the x-axis, and the impact force of the expected force response curve has a value between the average value and the peak value of the impact force of the actual force response curve. The present embodiment is set to 5.00 × 10⁴N, as shown in FIG. 3. In a coordinate system of the expected force response curve, an abscissa x is the displacement of the rigid impact plate, and the unit is m; the ordinate F is the collision force between the thin walled tube and the rigid impingement plate in units of N.

wherein minC is an objective function, F (x) is an actual force response curve, F⁰(x) To the desired force response curve, x₀And x_eRespectively the initial displacement and the final displacement of the rigid striking plate during the impact.

Step 5, determining design variables:

The thickness of each shell element of the 1/4 thin-walled tube in the finite element model is taken as the current design variable, namely the number of the design variables is the same as the number of the shell elements, and the total number of the shell elements is 1000 in the embodiment. The initial thickness of each shell element was 3 mm.

Step 6, determining constraint conditions:

using the thickness of the shell element and the mass of the 1/4 thin wall tube in the optimization process as constraints,

wherein t is a unit thickness matrix of the finite element model thin-walled tube; t is t_maxAnd t_minRespectively, the upper limit and the lower limit of the thickness of the shell element, in this embodiment, t_maxIs 5mm, t_minIs 0.5 mm; m is the mass of the 1/4 thin-wall pipe in the optimization process; m is₀1/4 the mass of a thin-walled tube with a thickness of 3 mm.

Step 7, obtaining a new finite element model

Obtaining a new finite element model by obtaining the value of the objective function of the 1/4 thin-wall tube optimization process in the step 4, specifically:

the value of the objective function of the 1/4 original finite element model is calculated using equation 4-1 in step 4.

Establishing the relation F (x) between the impact force of the 1/4 thin-wall pipe and the 1/4 rigid impact plate in the step 2 and the unit thickness,

is the thickness of the nth cell in the ith pass,

And (3) solving a formula 7-1 by using a constrained nonlinear multivariable optimization algorithm provided by MATLAB software to obtain the new thickness of each shell element in the thin-wall tube, and replacing the thickness of each shell element in the 1/4 original finite element model established in the step 1 by using the new thickness to obtain a new 1/4 finite element model.

Step 8, calculating the objective function value of the optimization process

A new value of the objective function is again obtained by means of the new 1/4 finite element model.

returning to the step 2, the method in the step 2 is used for obtaining the actual force response curve of the new 1/4 finite element model again.

And step 7, calculating a new objective function value through the actual force response curve. In calculating the new objective function value, the desired force response curve, the expression of the objective function, the design variables and the constraint conditions are the same as the conditions and parameters determined in steps 3 to 6. A new value of the objective function is obtained again and at the same time a new 1/4 finite element model is obtained again.

The actual force response curve of the new finite element model is again obtained using the new 1/4 finite element model obtained again in the iteration. Specifically, the step 2 is repeated to obtain the actual force response curve of the new 1/4 finite element model, the step 7 is entered, the iteration mode is repeated for 10 times, and 10 new objective function values are obtained in sequence, that is, one objective function value is obtained every iteration. Obtaining 11 objective function values in total, and numbering the objective function values as i in sequence; i is 1,2, … … 11.

I determining the minimum of 11 values of the objective function

C_i＝min[C₁,C₂,…,C₁₁]，

C_j＝min[C_i,C_i+1,…,C_i+10],j≥i，

if j ═ i, the iteration stops;

In this embodiment, the historical iteration curve of the objective function is as shown in fig. 4, and 23 iterations are performed in total to obtain 23 objective function values 10, and the minimum value 11 of the objective function is found in the 13 th iteration. When the value of the objective function is the smallest, the minimum and maximum thickness values of the shell element in the thin-walled tube are 2.44mm and 4.19mm, respectively. In the pure axial impact process of the thin-walled tube, the deformation mode of the thin-walled tube is checked through the after-treatment software LS-Prepost, the deformation condition before optimization is shown in figure 5, and the deformation condition after optimization is shown in figure 6, and the deformation condition is shown in the mode that the gradual buckling is formed from the front end to the rear end. However, two areas with thinner thickness appear at the front end of the optimized thin-walled tube, and the front end of the optimized thin-walled tube is similar to a trigger device and is locally bent when being impacted, so that the peak force is reduced.

Fig. 7 shows the actual force response curve 2 before optimization at an axial impact and the actual force response curve 3 after optimization at an axial impact. Compared with the curve 2, the curve 3 is closer to the expected force response curve and fluctuates around the expected force response curve, the average value of the response force of the curve 3 in the impact process is obviously higher than that of the curve 2, and the energy absorption capacity is stronger. The peak force, average force, impact load efficiency and specific energy absorption during impact for the thin walled tubes before and after optimization are shown in table 2 below.

Table 2:

	peak force N	Average force N	Efficiency of collision load	Specific energy absorption kJ/kg
					Before optimization	7.40×10⁴	3.33×10⁴	45.0％	9.6
After optimization	7.37×10⁴	4.71×10⁴	64.0％	10.8

Example 2

and establishing an original finite element model of the impacted thin-wall pipe according to the structural characteristics of the impacted thin-wall pipe and the rigid impact plate. As shown in fig. 8 and 9, the structural characteristics of the impacted thin-walled tube and the rigid impact plate are divided into two parts, one part is a thin-walled tube with a square cross section, the side length of the thin-walled tube is 100mm, the length of the thin-walled tube is 1000mm, and the wall thickness of the thin-walled tube is 3.0 mm; the other part is a rectangular rigid impact plate positioned at one end of the thin-walled tube, and the length of the impact plate is 880mm and the width of the impact plate is 400 mm.

The thin-wall pipe is positioned at one end of the rigid impact plate, and the geometric center line of the cross section of the thin-wall pipe is intersected with the center line of the rigid impact plate in the width direction. The thin-walled tube is inclined 15 degrees towards the short edge of the rigid impact plate, and an included angle alpha is formed between the end surface of the thin-walled tube and the surface of the rigid impact plate; the minimum distance delta between the end face of the thin-walled tube and the surface of the rigid impact plate is 5 mm. The end of the thin-wall pipe adjacent to the rigid impact plate is the front end.

The front end of the thin-walled tube is simulated to be impacted by the rigid impact plate. The impacted thin-wall pipe and the rigid impact plate are divided into two same parts in a shape like a Chinese character 'ri' along the length direction, the two parts are geometrically symmetrical, the impact force borne on the cross section of the thin-wall pipe when the thin-wall pipe is impacted is symmetrical, one part of the two parts is taken out, and the 1/2 thin-wall pipe and 1/2 rigid impact plate structure is obtained to be used as a geometric model for establishing an original finite element model of the thin-wall pipe under impact.

The thin-walled tube of 1/2 was divided into several shell elements by finite element software Hypermesh. To avoid hourglass deformation, all shell elements are fully integral shell elements. The rear end of the 1/2 thin-walled tube is fixed, so that the 1/2 rigid impact plate moves towards the 1/2 thin-walled tube at a constant speed of 5m/s, and the moving distance is 700 mm.

In this embodiment, each of the shell units has a size of 10 × 10mm and a wall thickness of 3.0mm, which is 2000 in total. The 1/2 rigid impact plate is a solid unit, 10 in total, as shown in FIG. 10.

The thin walled tube material properties are as in table 1.

Step 2, solving an actual force response curve of the structure:

the force response curve determined in this example is a force-displacement curve that reflects the relationship between the impact force between the thin walled tube and the rigid strike plate and the displacement of the rigid strike plate.

And (3) impacting the original finite element model established in the step (1) by utilizing a display dynamics analysis software LS-DYNA simulation, and obtaining a result file containing the impact process of the 1/2 rigid impact plate and the deformation of the 1/2 thin-walled tube by utilizing the display dynamics analysis software LS-DYNA. And (3) extracting impact force between the 1/2 thin-wall tube and the 1/2 rigid impact plate and displacement data of the rigid impact plate in a result file by using MATLAB software to obtain an actual force response curve of the 1/2 thin-wall tube under impact.

Step 3, setting an expected force response curve:

and obtaining the peak value and the average value of the actual force response curve of the thin-wall pipe when the thin-wall pipe is impacted by the MATLAB software. In this embodiment, the peak value of the impact force of the actual force response curve is 7.96 × 10⁴N, average 1.66X 10⁴And N is added. And (3) checking the deformation mode of the thin-walled tube by using a post-processing software LS-post. Distance of advance of rigid plate at the beginning of impactAt 0.1m, the 1/2 thin-wall pipe generates 2 fold deformations at the impacted end; as can be seen from the actual force response curve of the 1/2 thin-walled tube under impact, when the advancing distance of the rigid plate is within the range of 0-0.1 m, two larger peak forces appear in the actual force response, which are respectively 7.96 multiplied by 10⁴N and 7.49X 10⁴And N is added. After the advancing distance of the rigid plate is larger than 0.1m, Euler deformation occurs to the 1/2 thin-wall pipe, and the energy absorption effect is reduced; as can be seen from the actual force response curve of 1/2 thin-wall pipe when impacted, the actual force response drops sharply to 1.00X 10 after the advancing distance of the rigid plate is 0.1m⁴N or less until the end of the impact. In order to generate the progressive buckling deformation with high energy absorption and generate stable force response, the set expected force response curve comprises two parts: one part is a straight line with slope, the other part is a straight line parallel to the x-axis, and the maximum value of the straight line is in the range of 1.66 x 10⁴N～7.96×10⁴And N is added. The turning point of the two parts is between two larger peak forces, so as to reduce the magnitude of the first peak force and ensure that the force response is stable after 0.1 m. In this embodiment, the slope of the straight line with slope is 9.50 × 10 from the origin of coordinates⁵The maximum expected force value is 7.00X 10⁴N; the expected force value of a straight line parallel to the x-axis is 7.00X 10⁴N, as shown in FIG. 11. In a coordinate system of the expected force response curve, an abscissa x is the displacement of the rigid impact plate, and the unit is m; the ordinate F is the collision force between the thin walled tube and the rigid impingement plate in units of N.

wherein minC is an objective function, F (x) is an actual force response curve, F⁰(x) To the desired force response curve, x₀And x_eRespectively, during impactInitial displacement and final displacement of the striking plate.

Step 5, determining design variables:

The thickness of each shell element of the 1/2 thin-walled tube in the finite element model is taken as the current design variable, namely the number of the design variables is the same as the number of the shell elements, and the total number of the shell elements is 2000 in the embodiment. The initial thickness of each shell element was 3 mm.

Step 6, determining constraint conditions:

the thickness of the shell unit and the mass of the thin-walled tube in the optimization process are taken as constraint conditions,

wherein t is a unit thickness matrix of 1/2 thin-walled tubes in the finite element model, t_maxAnd t_minThe upper limit and the lower limit of the thickness of the shell unit are respectively 5mm and 0.5 mm; m is the mass of 1/2 thin-wall pipe in the optimization process, m ₀1/2 initial mass of thin wall tube.

Step 7, obtaining a new finite element model

the value of the objective function of the 1/2 original finite element model is calculated using equation 4-1 in step 4.

Establishing the relation F (x) between the impact force of the 1/2 thin-wall pipe and the 1/2 rigid impact plate in the step 2 and the unit thickness,

wherein, Delta E_kFor the change in kinetic energy during impact, Δ E_IIs the internal energy of the impact processIn the context of the variations of (a),

is the thickness of the nth cell in the ith pass,

And (3) solving a formula 7-1 by using a constrained nonlinear multivariable optimization algorithm provided by MATLAB software to obtain the new thickness of each shell element in the thin-wall tube, and replacing the thickness of each shell element in the 1/2 original finite element model established in the step 1 by using the new thickness to obtain a new 1/2 finite element model.

Step 8, calculating the objective function value of the optimization process

And obtaining a new objective function value through the new 1/2 finite element model.

And obtaining the new objective function value in an iterative mode. The initial value of the number of iterations is 11 due to the presence of numerical noise during the kinetic analysis. The method comprises the following steps:

returning to the step 2, the method in the step 2 is used for obtaining an actual force response curve of the new 1/2 finite element model.

And step 7, calculating a new objective function value through the actual force response curve. In calculating the new objective function value, the desired force response curve, the expression of the objective function, the design variables and the constraint conditions are the same as the conditions and parameters determined in steps 3 to 6. A new value of the objective function is obtained again and at the same time a new 1/2 finite element model is obtained again.

The actual force response curve of the new finite element model is again obtained using the new 1/2 finite element model obtained again in the iteration. Specifically, the step 2 is repeated to obtain the actual force response curve of the new 1/2 finite element model, the step 7 is entered, the iteration mode is repeated for 10 times, and 10 new objective function values are obtained in sequence, that is, one objective function value is obtained every iteration. Obtaining 11 objective function values in total, and numbering the objective function values as i in sequence; i is 1,2, … … 11.

I determining the minimum of 11 values of the objective function

C_i＝min[C₁,C₂,…,C₁₁]，

C_iIs the minimum of the 11 objective function values; i is the number of the minimum value; if i is 1, the iteration stops. If i is more than 1, continuing to iterate for i-1 time; in the process of iterating for i-1 times, obtaining i-1 objective function values; the total number of objective function values is 11+ i-1.

C_j＝min[C_i,C_i+1,…,C_i+10],j≥i，

if j ═ i, the iteration stops;

The historical iteration curve of the objective function is shown in fig. 12, 22 iterations are performed in total, 22 objective function values 10 are obtained, and the minimum value of the objective function is found at the 12 th iteration. When the value of the objective function is minimum, the minimum and maximum thickness values of the shell element in the thin-walled tube are 1.90mm and 3.86mm, respectively. In the process of oblique 15-degree impact of the thin-walled tube, the deformation condition before optimization is shown in fig. 13, in the initial stage of impact, the rigid plate advances by 0.1m, the thin-walled tube generates 2 fold deformations at the impacted end and has higher impact force, and after 0.1m, the thin-walled tube generates Euler deformation; the optimized deformation is shown in fig. 14 and is represented by progressive buckling from the anterior end to the posterior end. The front end of the optimized thin-walled tube becomes thinner, and the fixed end becomes thicker.

Fig. 16 shows the actual force response curve 5 before optimization for the desired force response broken line for oblique impacts and the actual force response curve 6 after optimization for the desired force response broken line for oblique impacts. The impact force after 0.1m of curve 5 drops sharply to 1.00X 10⁴And below N until the impact is finished, the curve 6 keeps a higher impact force level and fluctuates up and down near the expected force curve, the average value of the response force of the curve 6 in the impact process is obviously higher than that of the curve 5, and the energy absorption capacity is stronger. The peak force, average force, impact load efficiency and specific energy absorption during impact for the thin walled tubes before and after optimization are shown in table 3 below.

Table 3:

	peak force N	Average force N	Efficiency of collision load	Specific energy absorption kJ/kg
					Before optimization	7.96×10⁴	1.66×10⁴	20.8％	2.4
After optimization	8.71×10⁴	6.32×10⁴	72.6％	8.6

During the optimization under thin wall tube impact conditions described above, the values of the objective function, as shown in FIG. 12, appeared to be abrupt in the 6 th iteration, producing the low energy absorption Euler deformation as shown in FIG. 15.

Example 3

The thin walled tube material properties are as in table 1.

Step 2, solving an actual force response curve of the structure:

And (3) impacting the original finite element model established in the step (1) by utilizing a display dynamics analysis software LS-DYNA simulation, and obtaining a result file containing the impact process of the 1/2 rigid impact plate and the deformation of the 1/2 thin-walled tube by utilizing the display dynamics analysis software LS-DYNA. And (3) extracting impact force between the 1/2 thin-wall pipe and the 1/2 rigid impact plate and displacement data of the rigid impact plate in a result file by using MATLAB software to obtain an actual force response curve of the thin-wall pipe under impact.

Step 3, setting an expected force response curve:

and obtaining the peak value and the average value of the actual force response curve of the thin-wall pipe when the thin-wall pipe is impacted by the MATLAB software. In this embodiment, the peak value of the impact force of the actual force response curve is 7.96 × 10⁴N, average 1.66X 10⁴And N is added. And (3) checking the deformation mode of the thin-walled tube by using a post-processing software LS-post. At the initial stage of impact, the rigid plate advances by a distance of 0.1mThe 1/2 thin-wall pipe generates 2 fold deformations at the impacted end; as can be seen from the actual force response curve of the 1/2 thin-walled tube under impact, when the advancing distance of the rigid plate is within the range of 0-0.1 m, two larger peak forces appear in the actual force response, which are respectively 7.96 multiplied by 10⁴N and 7.49X 10⁴And N is added. After the advancing distance of the rigid plate is larger than 0.1m, Euler deformation occurs to the 1/2 thin-wall pipe, and the energy absorption effect is reduced; as can be seen from the actual force response curve of 1/2 thin-wall pipe when impacted, the actual force response drops sharply to 1.00X 10 after the advancing distance of the rigid plate is 0.1m⁴N or less until the end of the impact. In the results of example 2, as shown in fig. 12, the values of the objective function mutated in the 6 th iteration, resulting in the euler deformation with low energy absorption as shown in fig. 15. To avoid this during the optimization, the power function curve is set to the desired force response curve, in which the desired force response increases with increasing displacement, corresponding to a maximum response force of 7.50 × 10 at 0.7m⁴N, as shown in FIG. 17. In a coordinate system of the expected force response curve, an abscissa x is the displacement of the rigid impact plate, and the unit is m; the ordinate F is the collision force between the thin walled tube and the rigid impingement plate in units of N.

Step 5, determining design variables:

Step 6, determining constraint conditions:

Step 7, obtaining a new finite element model

is the thickness of the nth cell in the ith pass,

Step 8, calculating the objective function value of the optimization process

I determining the minimum of 11 values of the objective function

C_i＝min[C₁,C₂,…,C₁₁]，

C_j＝min[C_i,C_i+1,…,C_i+10],j≥i，

if j ═ i, the iteration stops;

As shown in fig. 18, the historical iteration curve of the objective function is obtained by performing 37 iterations in total to obtain 37 objective function values 10, and the minimum value of the objective function is found at the 27 th iteration. When the value of the objective function is minimum, the minimum and maximum thickness values of the shell element in the thin-walled tube are 0.82mm and 3.82mm, respectively. In the process of oblique 15-degree impact of the thin-walled tube, the deformation condition before optimization is shown in fig. 13, in the initial stage of impact, the rigid plate advances by 0.1m, the thin-walled tube generates 2 fold deformations at the impacted end and has higher impact force, and after 0.1m, the thin-walled tube generates Euler deformation; the optimized deformation is shown in fig. 19 and is represented by progressive buckling from the anterior end to the posterior end.

FIG. 20 shows the actual force response curve 8 before optimization for an oblique impact with the expected force response as a power function curve, and the oblique impactAnd when the lower expected force response is a power function curve, the optimized actual force response curve 9. The impact force of curve 8 after 0.1m drops sharply to 1.00X 10⁴Below N until the end of the impact, said curve 9 then increases gradually with the desired force curve, the average value of the response force of said curve 9 during the impact being significantly higher than that of said curve 8, the energy absorption capacity being also greater. The peak force, average force, impact load efficiency and specific energy absorption during impact for the thin walled tubes before and after optimization are shown in table 4 below.

Table 4:

	peak force N	Average force N	Efficiency of collision load	Specific energy absorption kJ/kg
					Before optimization	7.96×10⁴	1.66×10⁴	20.8％	2.4
After optimization	8.81×10⁴	5.45×10⁴	61.9％	8.4

The three examples described show that the present invention enables the actual force response curve of a thin walled tube to approach the desired force response curve. In an embodiment, the thin walled tube has an initial thickness of 3 mm. In example 1, the deformation modes before and after optimization under pure axial impact are progressive buckling; however, the optimized thin-wall pipe has better performance, and the four energy absorption indexes are improved.

In example 2 and example 3, two different objective functions are considered when the thin-walled tube is subjected to an impact of 15 degrees in an oblique direction. Under the action of oblique load, the thin-walled tube generates Euler deformation; in contrast, an optimized thin walled tube produces ideal progressive buckling.

Claims

1. An energy absorption structure optimization method taking expected force response course as a target is characterized by comprising the following specific processes:

according to the structural characteristics of the impacted thin-wall pipe and the rigid impact plate, obtaining a geometric model of the thin-wall pipe and the rigid impact plate, and establishing an original finite element model of the thin-wall pipe under impact;

simulating impact on the original finite element model established in the step 1 by utilizing display dynamics analysis software LS-DYNA, and obtaining an impact process of the rigid impact plate and a result file of thin-wall pipe deformation contained in the original finite element model; extracting impact force between the thin-walled tube and the rigid impact plate and data of displacement of the rigid impact plate in the result file by using MATLAB software to obtain an actual force response curve when the thin-walled tube is impacted;

the actual force response curve is a relationship between impact force and displacement; the impact force is the impact force between the thin-walled tube and the rigid impact plate, and the displacement is the displacement of the rigid impact plate;

step 3, setting an expected force response curve:

setting an expected force response curve by obtaining the peak value and the average value of the actual force response curve when the thin-walled pipe is impacted; the expected force response curve is the relationship between impact force and displacement set in the optimization process; the impact force is the impact force between the thin-walled tube and the rigid impact plate, and the displacement is the displacement of the rigid impact plate;

the expected force response curve is set to be a straight line parallel to the x-axis, and the average value of the impact force of the expected force response curve is 5.00 multiplied by 10⁴N; in a coordinate system of the expected force response curve, an abscissa x is the displacement of the rigid impact plate, and the unit is m; the ordinate F is the collision force between the thin-walled tube and the rigid impact plate, and the unit is N;

wherein minC is an objective function, F (x) is an actual response curve, F⁰(x) Is a target response force curve, x₀And x_eRespectively the initial displacement and the final displacement of the rigid impact plate in the impact process;

step 5, determining design variables:

the design variable is the thickness of each shell unit in the energy-absorbing structure optimization process; the optimization process is realized in an iterative mode; the thickness of each shell element changes in each iterative design;

taking the thickness of each shell unit of the thin-walled tube in the original finite element model as a current design variable respectively, wherein the number of the design variables is the same as that of the shell units; the initial thickness of each shell element is 3 mm;

step 6, determining constraint conditions:

wherein t is a unit thickness matrix of the finite element model thin-walled tube; t is t_maxAnd t_minUpper and lower limits for the thickness of the shell element, respectively; m is the mass of the 1/4 thin-wall pipe in the optimization process; m is₀1/4 the mass of a thin-walled tube with a thickness of 3 mm;

step 7, obtaining a new finite element model:

calculating an objective function value of the original finite element model by using a formula 4-1 in the step 4;

is the thickness of the nth cell in the ith pass,

is the thickness of the nth cell in the (i-1) th time, and N is the number of the cells; x is the initial displacement x of the rigid impact plate during impact₀Final displacement x from rigid impact plate_eAny position displacement therebetween;

solving a formula 7-1 by using a constrained nonlinear multivariable optimization algorithm provided by MATLAB software to obtain the new thickness of each shell unit in the thin-walled tube, and replacing the thickness of each shell unit in the original finite element model established in the step 1 by using the new thickness to obtain a new finite element model;

step 8, calculating an objective function value of the optimization process:

obtaining a new objective function value again through the new finite element model;

obtaining the new objective function value again in an iterative mode; the initial value of the iteration times is 11 times due to the fact that numerical noise exists in the dynamic analysis process; the method comprises the following steps:

returning to the step 2, obtaining the actual force response curve of the new finite element model again by using the method in the step 2;

step 7, calculating a new objective function value through the actual force response curve; in the calculation of the new objective function value, the required expected force response curve, the expression of the objective function, the design variables and the constraint conditions are the same as the conditions and parameters determined in the steps 3 to 6; obtaining a new objective function value again, and obtaining a new finite element model again at the same time;

continuously obtaining an actual force response curve of the new finite element model by using the new finite element model obtained again in the iteration;

returning to the step 2 to obtain an actual force response curve of the new finite element model, entering the step 7, repeating the iteration mode for 10 times, and sequentially obtaining 10 new objective function values, namely obtaining one objective function value every time iteration is performed; obtaining 11 objective function values in total, and numbering the objective function values as i in sequence; 1,2, … … 11;

i determining the minimum of 11 values of the objective function

C_i＝min[C₁,C₂,…,C₁₁]，

C_iIs the minimum of the 11 objective function values; i is the number of the minimum value; if i is 1, the iteration stops; if i is more than 1, continuing to iterate for i-1 time; in the process of iterating for i-1 times, obtaining i-1 objective function values; obtaining 11+ i-1 objective function values in total;

II, taking the minimum value C of the ith, i +1, … i +10 in the 11+ i-1 objective function values_jJ is the minimum valueNumbering; the method comprises the following steps:

C_j＝min[C_i,C_i+1,…,C_i+10],j≥i，

if j ═ i, the iteration stops;

if j is larger than i, continuing to iterate j-i times, and obtaining j-i objective function values in the process of iterating j-i times;

obtaining 11+ i-1+ j-i objective function values in total;

III, repeating the process of obtaining the j-i objective function value in the II until the minimum value of the obtained current 11 objective functions is equal to the first value of the current 11 objective functions, and stopping iteration; if the iteration times reach 50 times, the minimum value in the current 11 objective functions still cannot be equal to the first value in the current 11 objective functions, and the iteration is also stopped; so far, the optimization of the energy absorption structure is completed.

2. The method for optimizing an energy absorbing structure with the aim of expecting a force response course as recited in claim 1, wherein the structural characteristics of the impacted thin-walled tube and the rigid impact plate are divided into two parts, one part is the thin-walled tube with a square cross section, and the other part is the rectangular rigid impact plate; the rigid impact plate is positioned at one end face of the thin-walled tube, and the central line of the thin-walled tube in the length direction is superposed with the geometric center of the rigid impact plate, or the geometric central line of the cross section of the thin-walled tube is superposed with the central line of the rigid impact plate in the width direction; the minimum distance between the end face of the thin-walled tube and the surface of the rigid impact plate is 5 mm.

3. The method for optimizing an energy absorbing structure with the objective of an expected force response history as claimed in claim 2, wherein when the center line of the thin-walled tube in the length direction coincides with the geometric center of the rigid impact plate, the impacted thin-walled tube and the rigid impact plate are equally divided into four parts along the length direction, and one of the four parts is taken to obtain 1/4 thin-walled tube and rigid impact plate structures as a geometric model for building an original finite element model of the thin-walled tube under impact;

4. The method for optimizing an energy absorbing structure with the aim of expecting a force response course as claimed in claim 2, wherein the obtained thin-walled tube in the geometric model of 1/4 or the thin-walled tube in the geometric model of 1/2 is divided into a plurality of fully integrated shell units through finite element software Hypermesh; dividing the rigid impact plate in the geometric model of 1/4 or 1/2 into several solid units.

5. The method of claim 1, wherein the desired force response curve is a straight line parallel to the x-axis and the expected force response curve has an average impact force of 5.00 x 10⁴N; or a broken line divided into two parts, one part is a straight line with a slope, the other part is a straight line parallel to the x-axis, and the slope of the straight line with the slope is 9.50 multiplied by 10 from the coordinate origin⁵The maximum expected force value is 7.00X 10⁴N; the expected force value of a straight line parallel to the x-axis is 7.00X 10⁴N; or a power function curve in which the expected force response increases with increasing displacement, corresponding to a maximum response force of 7.50 x 10 at 0.7m⁴N; in a coordinate system of the expected force response curve, an abscissa x is the displacement of the rigid impact plate, and the unit is m; the ordinate F is the collision force between the thin walled tube and the rigid impingement plate in units of N.

6. The method for optimizing an energy absorbing structure with a view to a desired force response history according to claim 1, wherein in step 6, the upper limit t of the thickness of the shell element is set_max5mm, lower limit t of the thickness of the shell element_min＝0.5mm。

7. The method of claim 1, wherein the peak force of the optimized energy absorbing structure is 7.37 to 8.81 x 10⁴N, average force of 4.71-6.32 x 10⁴N, the collision load efficiency is 61.9-72.6%, and the specific energy absorption is 8.4-10.8 kJ/kg.