CN103218500A - Robust automobile body material matching method based on multi-objective continuous orthogonal method - Google Patents

Abstract

The invention discloses a robust automobile body material matching method based on a multi-objective continuous orthogonal method. The method mainly comprises the following steps: creating a collision finite element model through actual collision states and conditions; and then, determining an optimization target, constraint conditions and design variables to be optimized, and creating a three-level orthogonal table; obtaining target response values through numerical simulation calculation, calculating target signal to noise ratios and obtaining correction response values; and finally, reasonably selecting an optimal solution through the determined comprehensive evaluation value, judging the designed target satisfies an iteration judging condition, if not, using the current optimal solution as the initial solution of the next iteration step until the iteration judging condition is satisfied. According to the invention, automobile body design can be further guided reasonably, and the automobile safety performance and lightweight design level are improved.

Description

A kind of robustness body of a motor car match materials method based on the continuous orthogonal method of multiple goal

Technical field

The present invention relates generally to body of a motor car lightweight and security fields, refers in particular to a kind of robustness body of a motor car match materials method based on the continuous orthogonal method of multiple goal.

Background technology

Along with developing of automobile industry, auto industry requires more and more higher to aspects such as lightweight, safety.Simultaneously, the automobile lightweight has also obtained the great attention of countries in the world as the important channel of energy-saving and emission-reduction.In the automobile weight reduction process, body of a motor car accounts for about the 30%-60% of car load gross weight, and about 70% fuel is to consume on body quality in the car running process, so the vehicle body loss of weight is one of light-weighted most important approach of automobile.For the vehicle body loss of weight, often adopt number of ways such as lightweight structure and lightweight material to reduce body quality on the engineering, feasible deep day by day for the research of lightweight material, increasing material has used in the body of a motor car light-weight design process.Therefore, guaranteeing under the condition that body of a motor car crashworthiness performance does not descend that it is significant how to make that by reasonably combined material the body of a motor car quality descends.

Body of a motor car match materials process is non-linear a, multivariate and multiobject optimal design process, need satisfy under the particular constraints condition, seeks the crash-worthiness and the lightweight index optimum of body of a motor car.In traditional method for designing, the method that adopts component materials constantly to replace obtains suitable material, and this process need passes through constantly to revise and trial and error could obtain suitable match materials, and is not suitable for the more situation of material number.Simultaneously, successively single part is carried out material and replace, ignored influencing each other between each parts.

Though along with the continuous development of discrete optimization, adopt genetic algorithm such as NGSA-II to separate, choose optimum solution according to designing requirement again by the Pareto that obtains corresponding problem after will the discrete variable serialization.It is too much that but these class methods are called the direct problem number of times, and excessive for large-scale calculations problem calculated amount, computing time is of a specified duration, has lost engineering significance.And adopting the agent model method to replace direct problem, the fitting precision of agent model directly influences the optimization result, and agent model can not react the characteristic of each material itself really.In body of a motor car match materials process, find a kind of method fast and effectively to reach body of a motor car material optimal design, extremely important for the slip-stick artist.

Summary of the invention

The objective of the invention is to solve traditional material replacement method poor accuracy, having ignored reciprocal effect between the parts and genetic algorithm, to call the direct problem number of times too much, problem such as agent model method precision is low.This method can effectively be applied to match materials optimal design etc. under each impact conditions.

The present invention solves the technical method that its technical matters adopts: a kind of robustness body of a motor car match materials method based on the continuous orthogonal method of multiple goal comprises the steps:

Step 1: determine body of a motor car match materials optimal design impact conditions, this impact conditions comprises head-on crash, side collision and offset collision etc.;

Step 2: set up corresponding finite element model by pre-processing software HyperMesh, and the validity by this model of verification experimental verification;

Step 3: at corresponding laws and regulations and engineering experience, determine optimal design target, constraint under this impact conditions, and definite body of a motor car match materials design variable and each design variable space;

Step 4: adopt three horizontal quadrature tables to be optimized design, determine the initial designs level, the method for determining is: Level2 is an initial designs, and Level1 and Level3 are respectively the adjacent chosen candidate value of Level2.Determine corresponding three horizontal quadrature tables at the design variable number;

Step 5: adopt the Ls-dyna of simulation software that each sample point of orthogonal arrage is carried out simulation calculation, obtain the simulation value of each response.Calculate the signal to noise ratio (S/N ratio) of each target according to simulation value, its concrete computing method are:

Hope little characteristic: $S/N=-10\log \left(\frac{\mathrm{\Σ}{y}_i^2}{n}\right)$

Hope big characteristic: $S/N=-10\log \left(\frac{\mathrm{\Σ}1/y_i^2}{n}\right)$

Hope the order characteristic: $S/N=-10\log \left(\frac{{\stackrel{\‾}{y}}^2}{s^2}\right)$

In the formula: n is a test number (TN), y _iBe the response of the i time of each target test,

Be the mean value of the n time test, s ²Be variance.

Adopt penalty function method that each binding occurrence is attached on the signal to noise ratio (S/N ratio) of each target, obtain the correction response of each target, concrete computing method are:

R _new＝S/N+P(x)

$P\left(x\right)=s\×\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\max [0,v_i]$

In the formula: S/N is a signal to noise ratio (S/N ratio), R _NewFor revising response, P (x) is a penalty, v _iBe the penalties of i constraint, s is a penalty factor.

And obtain the comprehensive grading value of each sample point by weighted scoring value method, because signal to noise ratio (S/N ratio) all requires maximum, so the weighted scoring value calculating method is:

$Y_j=\underset{i=1}{\overset{k}{\mathrm{\Σ}}}{b}_i\frac{y_{\mathrm{ji}}}{y_{0i}}$

In the formula: j is a tested number; I is the index sequence number; b _iBe the weights of i index; Y _jIt is the comprehensive grading of the j time test; y _JiIt is the i item desired value of the j time test; y _0iI item desired value for reference test; K is the index number;

Step 6: adopt the ANOM method promptly to calculate the mean value of test findings sum of 1,2,3 level correspondences of each factor, obtain the optimal design result of this iteration step;

Step 7: judge whether design object satisfies (1) continuous 5 optimal design results and do not improve; (2) iterations reaches maximum, and the acquiescence greatest iteration number is that 2nmax(nmax is maximum alternative level sum).Then obtain final design result if satisfy condition, then each variable design space is not moved if do not satisfy condition, the initial designs of determining the n+1 time orthogonal design is the optimal result of the n time orthogonal design, and gets back to step 4.

The invention has the beneficial effects as follows:

1, adopts the continuous orthogonal method of multiple goal of the present invention's proposition, can obtain the optimal design result of body of a motor car match materials.Gained optimal design result can guarantee that the crash-worthiness condition is issued to the lightweight purpose, reduces body of a motor car lightweight cost and cycle.

2, the continuous orthogonal method computation process of multiple goal of the present invention's proposition is directly called direct problem, therefore can react each properties of materials preferably.Simultaneously, compare the large-scale calculations problem that genetic algorithm, agent model method etc. can better solve the body of a motor car match materials because its iterations is few and be applicable to characteristic such as multiple goal.

3, consider the interactivity of material and thickness in the design process, and material and thickness have been carried out Robust Optimization, made the optimal design result have better robustness.

Description of drawings

Fig. 1 is a kind of robustness body of a motor car material based on the continuous orthogonal method of multiple goal of the present invention

The matching process process flow diagram.

Fig. 2 is certain vehicle head-on crash finite element model.

Fig. 3 is a head-on crash match materials design part.

Embodiment

Be that example is elaborated with certain vehicle frontal crash of vehicles crash-worthiness match materials below.

Fig. 1 shows the process flow diagram of a kind of robustness body of a motor car match materials method based on the continuous orthogonal method of multiple goal of the present invention.A kind of robustness body of a motor car match materials method based on the continuous orthogonal method of multiple goal of the present invention comprises the steps:

Step 1: determine that this vehicle match materials purpose of design designs for head-on crash.

Step 2: adopt Hypermesh to set up the head-on crash finite element model of this vehicle.As shown in Figure 2.And determine that according to C-NCAP impact velocity is 50km/h.

Step 3: according to existing 2012 editions C-NCAP, " member protection of GB11551-2003 passenger car head-on crash " and designer's preference; consider the body of a motor car lightweight simultaneously; selected optimal design target and corresponding weight value are as shown in table 1; constraint condition is that the amount of moving is not more than 77mm on the steering column, after the amount of moving be not more than 90mm.

Table 1:

Choose that automobile component is as body of a motor car crash-worthiness match materials design part as shown in Figure 3, the 1-3 parts for example are respectively subframe, subframe arm, front longitudinal.X ₁_ X ₃Choosing alternative materials according to demand for each component materials is respectively: ST12, ST14, ST16, DP500, DP590, TR590, DP780.X ₄_ X ₆Choosing alternative thickness according to demand for each component thickness is respectively: 1.0,1.2,1.4,1.6,1.8,2.0,2.2,2.4,2.6.

Step 4: determine that the orthogonal design level is respectively: Level2 is an initial designs, and Level1 and Level3 are respectively the adjacent chosen candidate value of Level2.Because design variable always has 7, chooses L ₁₈(3 ⁷) orthogonal arrage, this orthogonal arrage comprises 18 sample point 7 factors, and each factor has three levels;

Step 5: adopt the Ls-dyna of simulation software that each sample point of orthogonal arrage is carried out simulation calculation, obtain the simulation value of each response.Calculate the signal to noise ratio (S/N ratio) of each target according to simulation value, its concrete computing method are:

Hope little characteristic: $S/N=-10\log \left(\frac{\mathrm{\Σ}{y}_i^2}{n}\right)$

Hope big characteristic: $S/N=-10\log \left(\frac{\mathrm{\Σ}1/y_i^2}{n}\right)$

Hope the order characteristic: $S/N=-10\log \left(\frac{{\stackrel{\‾}{y}}^2}{s^2}\right)---\left(1\right)$

In the formula (1): n is a test number (TN), y _iBe the response of the i time of each target test,

Be the mean value of the n time test, s ²Be variance.

Hope little characteristic because the design object of certain vehicle frontal crash of vehicles crash-worthiness match materials problem is, therefore only need adopt formula (2):

$S/N=-10\log \left(\frac{\mathrm{\Σ}{y}_i^2}{n}\right)---\left(2\right)$

Adopt penalty function method that each binding occurrence is attached to each target signal to noise ratio, obtain the correction response of each target, concrete computing method are:

R _new＝S/N+P(x)

$P\left(x\right)=s\×\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\max [0,v_i]---\left(3\right)$

In the formula (3): S/N is a signal to noise ratio (S/N ratio), R _NewFor revising response, P (x) is a penalty, v _iBe the penalties of i constraint, s is a penalty factor.

And obtain the comprehensive grading value of each sample point by weighted scoring value method, because signal to noise ratio (S/N ratio) all requires maximum, so the weighted scoring value calculating method is:

$Y_j=\underset{i=1}{\overset{k}{\mathrm{\Σ}}}{b}_i\frac{y_{\mathrm{ji}}}{y_{0i}}---\left(4\right)$

In the formula (4): j is a tested number; I is the index sequence number; b _iBe the weights of i index; Y _jIt is the comprehensive grading of the j time test; y _JiIt is the i item desired value of the j time test; y _0iI item desired value for reference test; K is the index number.

See Table 2 by the comprehensive grading value that calculates each test.

Table 2:

Tested number	X 1	X 2	X 3	X 4	X 5	X 6	The comprehensive grading value
								1	1	1	1	1	1	1	α 1
2	1	1	2	2	3	3	α 2
								3	1	2	1	3	3	2	α 3
4	1	2	3	1	2	3	α 4
								5	1	3	2	3	2	1	α 5
6	1	3	3	2	1	2	α 6
								7	2	1	1	3	2	3	α 7
8	2	1	3	1	3	2	α 8
								9	2	2	2	2	2	2	α 9
10	2	2	3	3	1	1	α 10
								11	2	3	1	2	3	1	α 11

12	2	3	2	1	1	3	α 12
								13	3	1	2	3	1	2	α 13
14	3	1	3	2	2	1	α 14
								15	3	2	1	2	1	3	α 15
16	3	2	2	1	3	1	α 16
								17	3	3	1	1	2	2	α 17
18	3	3	3	3	3	3	α 18

Step 6: adopt the mean value of ANOM method by the test findings sum of 1,2,3 level correspondences of each factor of calculating, as

$X_{12}=\frac{{\mathrm{\α}}_7+{\mathrm{\α}}_8+{\mathrm{\α}}_9+{\mathrm{\α}}_{10}+{\mathrm{\α}}_{11}+{\mathrm{\α}}_{12}}{6},$ $X_{13}=\frac{{\mathrm{\α}}_{13}+{\mathrm{\α}}_{14}+{\mathrm{\α}}_{15}+{\mathrm{\α}}_{16}+{\mathrm{\α}}_{17}+{\mathrm{\α}}_{18}}{6},$

X wherein ₁₁, X ₁₂, X ₁₃, X ₂₁Represent the 1st parts the 1st, 2,3 levels and the 2nd the pairing mean value of parts the 1st level respectively.Choose the optimal design result of the level of its intermediate value maximum for this iteration step.

Step 7: judge whether design object satisfies (1) continuous 5 optimal design results and do not improve; (2) iterations reaches maximum, and the acquiescence greatest iteration number is 12.Then obtain optimal result if satisfy condition, then each variable design space is not moved if do not satisfy condition, the design space moving method is that the initial designs of the n+1 time orthogonal design is the optimal level of the n time orthogonal design, and returns step 4 beginning next round orthogonal design.

Below only be the preferred embodiments of the invention, protection scope of the present invention also not only is confined to above-mentioned example, and all technical schemes that belongs under the thinking of the present invention all are considered as protection scope of the present invention.Should be pointed out that for those skilled in the art the some improvements and modifications not breaking away under the principle of the invention should be considered as protection scope of the present invention.

Claims (10)

1. the body of a motor car match materials method based on the continuous orthogonal method of multiple goal is characterized in that, comprises the steps:

Step 1: determine body of a motor car match materials optimal design impact conditions;

Step 2: set up corresponding finite element model, and the validity by this model of verification experimental verification;

Step 3: at corresponding collision rules and engineering experience, determine optimal design target and constraint under this impact conditions, and the value space of definite body of a motor car match materials design variable and each design variable;

Step 4: adopt three horizontal quadrature tables to be optimized design, determine initial designs level and corresponding three horizontal quadrature tables at the design variable number;

Step 5: each sample point of orthogonal arrage is carried out simulation calculation, obtain the simulation value of each response, and calculate the signal to noise ratio (S/N ratio) of each target, adopt penalty function method that each binding occurrence is appended to each target signal to noise ratio, obtain the correction response of each target, and by calculating the comprehensive grading value of each sample point, poly-by this step, the multi-objective restriction optimization problem is transformed for single goal unconfinement optimization problem;

Step 6: the optimal design result who obtains this iteration step;

Step 7: judge whether design object satisfies the iteration decision condition,, then each variable design space is not moved if do not satisfy condition, and get back to step 4 if satisfying the iteration decision condition then obtains final design result.

2. a kind of body of a motor car match materials method based on the continuous orthogonal method of multiple goal as claimed in claim 1, it is characterized in that: the body of a motor car match materials optimal design impact conditions that described step 1 comprised comprises head-on crash, offset collision, side collision.

3. a kind of body of a motor car match materials method based on the continuous orthogonal method of multiple goal as claimed in claim 1 is characterized in that: utilize pre-processing software Hypermesh to set up described finite element model in the described step 2.

4. a kind of robustness body of a motor car match materials method as claimed in claim 1 based on the continuous orthogonal method of multiple goal, it is characterized in that: in the described step 4 in the three horizontal quadrature tables three level values of each design variable make following regulation, that is: Level2 is an initial designs, and Level1 and Level3 are respectively the adjacent chosen candidate value of Level2.

5. a kind of body of a motor car match materials method based on the continuous orthogonal method of multiple goal as claimed in claim 1 is characterized in that: described step 5 adopts business software LS-DYNA to carry out simulation calculation and obtains the simulation value of each response.

6. a kind of body of a motor car match materials method as claimed in claim 5 based on the continuous orthogonal method of multiple goal, it is characterized in that: the concrete computing method of the signal to noise ratio (S/N ratio) of each target are in the described step 5:

Hope little characteristic: $S/N=-10\log \left(\frac{\mathrm{\Σ}{y}_i^2}{n}\right)$

Hope big characteristic: $S/N=-10\log \left(\frac{\mathrm{\Σ}1/y_i^2}{n}\right)$

Hope the order characteristic: $S/N=-10\log \left(\frac{{\stackrel{\‾}{y}}^2}{s^2}\right)$

In the formula: n is a test number (TN), y _iBe the response of the i time test,

Be the mean value of the n time test, s ²Be variance.

7. a kind of body of a motor car match materials method as claimed in claim 1 based on the continuous orthogonal method of multiple goal, it is characterized in that: described step 5 is attached to each target signal to noise ratio by adopting penalty function method with each binding occurrence, obtain the correction response of each target, concrete computing method are:

R _new＝S/N+P(x)

$P\left(x\right)=s\×\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\max [0,v_i]$

In the formula: S/N is a signal to noise ratio (S/N ratio), R _NewFor revising response, P (x) is a penalty, v _iBe the penalties of i constraint, s is a penalty factor.

8. a kind of body of a motor car match materials method as claimed in claim 1 based on the continuous orthogonal method of multiple goal, it is characterized in that: described step 5 adopts weighted scoring value method to obtain the comprehensive grading value of each sample point, because each signal to noise ratio (S/N ratio) requires to be maximum, so the weighted scoring value calculating method is

$Y_j=\underset{i=1}{\overset{k}{\mathrm{\Σ}}}{b}_i\frac{y_{\mathrm{ji}}}{y_{0i}}$

In the formula: j is a tested number; I is the index sequence number; b _iBe the weights of i index; Y _jIt is the comprehensive grading of the j time test; y _JiIt is the i item desired value of the j time test; y _0iI item desired value for reference test; K is the index number.

9. a kind of body of a motor car match materials method as claimed in claim 1 based on the continuous orthogonal method of multiple goal, it is characterized in that: optimal design result's the employing ANOM method of choosing under this iteration step in the described step 6, its computing method are the mean value of the test findings sum of 1,2,3 level correspondences of each factor of calculating.

10. a kind of body of a motor car match materials method as claimed in claim 1 based on the continuous orthogonal method of multiple goal, it is characterized in that: the iteration decision condition of described step 7 is: (1) continuous 5 optimal design results do not improve; (2) iterations reaches maximum, and the acquiescence greatest iteration number is 2nmax, and wherein nmax is maximum alternative level sum.

Images