CN103984838B - A kind of method determining side airbag shape - Google Patents
A kind of method determining side airbag shape Download PDFInfo
- Publication number
- CN103984838B CN103984838B CN201410242898.3A CN201410242898A CN103984838B CN 103984838 B CN103984838 B CN 103984838B CN 201410242898 A CN201410242898 A CN 201410242898A CN 103984838 B CN103984838 B CN 103984838B
- Authority
- CN
- China
- Prior art keywords
- optimization
- air bag
- bladder shape
- shape variable
- dummy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Air Bags (AREA)
Abstract
The invention discloses a kind of method determining side airbag shape, including selecting to optimize point, using the position coordinates of optimization point as bladder shape variable；Obtain similar area model and the approximation damage model of functional relationship between reflection bladder shape variable and air bag area and dummy's damage data；Bladder shape variable is optimized sampling；By in the optimization bladder shape variable input similar area and the damage model that optimize sampling acquisition, it is thus achieved that approximation air bag area and approximation dummy's damage data；Choose so that the optimization bladder shape variable approximating the minimum predetermined number of air bag area is made to optimize solution making approximation dummy's damage data meet in the optimization bladder shape variable of damage criterion constraints；Obtain and solve corresponding air bag area and dummy's damage data with optimizing；Choose so that the optimization solution of air bag area minimum makees target solution making dummy's damage data meet in the optimization solution of damage criterion constraints.The present invention less calculating can measure more accurate optimum results.
Description
Technical field
The present invention relates to automobile technical field, particularly relate to a kind of determination side based on geometric parameter
The method of air bag shape.
Background technology
Along with the fast development of automobile industry, vehicle passive safety is increasingly paid attention to by people.Safety,
Energy-conservation and environmental protection also becomes three big design motives of current automotive engineering field.Automobile side is touched
For hitting, side airbag protects occupant beyond doubt, it is to avoid occupant and car body inside gadget generation secondary
The most important safety aid of collision, its effect in the passenger protection of side is by auto industry
Boundary generally acknowledges, has had become as the basic security configuration of many vehicles, particularly chest and abdomen side airbag
Air bag can protect easily vulnerable chest and the abdominal sites of occupant in side collision effectively.For
The design of side airbag, the design of its bladder shape is again the emphasis in design and difficult point, because
The shape design of side airbag is influenced by many factors, including occupant side space of planes, door interior trim
Plate moulding, door internal decoration plate rigidity, car body invasion speed and invasion amount, car body deformable contour, seat becomes
The many factors such as shape and passengers motion attitude, therefore in actual design, the shape of side airbag is past
Toward needing repeatedly to revise, to realize preferably passenger protection effect.
Determine that the method for side airbag shape mainly has two kinds: one to be test trial-and-error method, i.e. at present
By manual modification bladder shape repeatedly, side sled test contrast occupant injury situation is repeated and comes
Realize final Shape optimization；Another kind is emulation trial-and-error method, by manual modification bladder shape repeatedly,
Repeatedly running occupant injury analysis model and optimize bladder shape, this occupant injury analyzes model e.g.
The MADYMO (Mathematical Dynamic Model) of TNO company.Test trial-and-error method is with imitative
True trial-and-error method both approaches is all according to engineering experience for the optimization design of bladder shape, hands repeatedly
Dynamic amendment bladder shape, repeatedly trial and error, the sampling that can carry out due to manual modification is limited, Er Qiezhen
The design of bladder shape is lacked again clear and definite method as guidance, therefore it is difficult to ensure that bladder shape one
The optimization of secondary property puts in place, it is also difficult to obtains the passenger protection effect of optimum, considerably increases side airbag gas
The development cost of capsule and construction cycle, be therefore badly in need of a kind of science, efficiently, full automation and
The method of the determination side airbag shape of Performance optimization.
Summary of the invention
The purpose of the embodiment of the present invention is to solve development cost height and the construction cycle that existing method exists
Long problem, it is provided that a kind of method determining side airbag shape efficiently.
For achieving the above object, the technical solution used in the present invention is: one determines side airbag
The method of shape, including:
Select to optimize point, all points that optimize relatively are optimized the position coordinates of coordinate system determine as being used for
The bladder shape variable of side airbag shape；
Obtain training bladder shape variable, and bladder shape variable and air bag area and dummy are used in training
Corresponding relation between damage data；
Utilize the corresponding relation between training bladder shape variable and air bag area, obtain reflection air bag
The similar area model of functional relationship between configuration variable and air bag area；
Utilize the corresponding relation between training bladder shape variable and dummy's damage data, it is thus achieved that reflection
The approximation damage model of functional relationship between bladder shape variable and dummy's damage data；
Described bladder shape variable is optimized sampling, it is thus achieved that optimize and use bladder shape variable；
Optimization bladder shape variable is separately input into described similar area model and approximation damage model
In, to obtain the approximation air bag area corresponding with optimization bladder shape variable and approximation dummy respectively
Damage data；
Become in the optimization bladder shape making approximation dummy's damage data meet damage criterion constraints
Amount is chosen the optimization bladder shape variable of the predetermined number minimum so that approximation air bag area as excellent
Dissolve；
Obtain and solve corresponding air bag area and dummy's damage data with optimizing；
Choose so that air bag making dummy's damage data meet in the optimization solution of damage criterion constraints
The optimization solution of area minimum is as target solution；
Side airbag shape is determined according to described target solution.
Preferably, described method also includes:
Described optimization solution is increased in described training bladder shape variable, and to described similar area
Model and approximation damage model are trained updating；
Described described bladder shape variable be optimized sampling include:
Described bladder shape variable is optimized sampling, until meeting iteration stopping condition.
Described choose in the optimization solution making dummy's damage data meet damage criterion constraints so that
The optimization solution of air bag area minimum includes as target solution:
Selecting through last optimization sampling of damage criterion constraints is met making dummy's damage data
In the optimization solution taken, choose the optimization solution so that air bag area minimum as target solution.
Preferably, described method also includes:
Supplementary training sampling is carried out near solution, it is thus achieved that training becomes with supplementing bladder shape described optimization
Amount；
Obtain with training with supplementing the corresponding air bag area of bladder shape variable and dummy's damage data；
Described optimization is solved and training increases to described training bladder shape with supplementary bladder shape variable
In variable, and it is trained updating to described similar area model and approximation damage model.
Preferably, described iteration stopping condition is: described bladder shape variable is optimized sampling
Number of times reach default largest optimization sampling number, or adjacent two suboptimization are sampled the target chosen
Solve the difference of corresponding air bag area less than or equal to preset value.
Preferably, described selection optimization point includes:
Determine the original shape of side airbag shape and relatively optimize the initial position of coordinate system；
The profile of described original shape selects optimize point.
Preferably, described acquisition training bladder shape variable, and training with bladder shape variable with
Corresponding relation between air bag area and dummy's damage data includes:
Described bladder shape variable is trained sampling, to obtain training bladder shape variable；
Generate the side airbag finite element analysis mould corresponding with described training bladder shape variable
The grid of type, as training grid；
Utilize described training grid, obtain training bladder shape variable and air bag area and dummy's damage
Corresponding relation between data；And/or,
Air bag area and dummy's damage data that described acquisition is corresponding with optimizing solution include:
Generate and the described grid optimizing the corresponding side airbag finite element analysis model of solution, make
For optimizing grid；
Utilize described optimization grid, obtain and solve corresponding air bag area and dummy's damage data with optimizing.
Preferably, described described bladder shape variable is trained sampling includes: use test to set
Meter method is trained sampling to described bladder shape variable；And/or,
Described described bladder shape variable is optimized sampling includes: use test design method to institute
State bladder shape variable and be optimized sampling.
Preferably, the optimization balloon-shaped described bladder shape variable being optimized sampling and obtain
The group number of shape variable is at least and described bladder shape variable is trained sampling and obtains training air bag
10 times of the group number of configuration variable.
Preferably, described utilize described training grid, obtain training bladder shape variable and air bag
Corresponding relation between area and dummy's damage data includes:
Described training grid is separately input into air bag areal calculation model and occupant injury analyzes model
In, to obtain the air bag area corresponding with training bladder shape variable and dummy's damage data respectively；
And/or,
Described utilize described optimization grid, obtain and solve corresponding air bag area and dummy's damage with optimizing
Data include:
Described optimization grid is separately input into air bag areal calculation model and occupant injury analyzes model
In, solve corresponding air bag area and dummy's damage data to obtain respectively with optimizing.
Preferably, described optimization coordinate system is two-dimensional coordinate system, the first seat of described optimization coordinate system
Parameter extends along the direction of backrest, and the second coordinate axes of described optimization coordinate system is along length of wagon
Direction extend perpendicular on vertical with described first coordinate axes.
Preferably, described original shape includes that the limit parallel with described first coordinate axes is pacified as side
The installation limit of full air bag, is positioned on the direction of described second coordinate axes and installs the limit in front, limit as excellent
Changing limit, connect two straightways optimizing limit with installing limit, described optimization point is positioned at the four of original shape
On individual angle point and on described optimization limit.
The beneficial effects of the present invention is, the method for the determination installing side airbag shape of the present invention is primarily based on
Between training bladder shape variable acquisition reflection bladder shape variable and air bag area, functional relationship is near
Damage like the approximation of functional relationship between Area Model, and reflection bladder shape variable and dummy's damage data
Wound model, owing to this approximate model calculates based on functional relationship, and and non-real phantom,
Therefore, it can the optimization bladder shape variable according to input and quickly calculate approximation air bag area with near
Like dummy's damage criterion, be optimized counting of sampling can be very big, to filter out more every time
Optimize solution accurately, be such as input into the optimization solution filtered out in phantom the most again carrying out really
Recognize, and then relatively small amount of calculation can obtain more accurate optimum results.
Accompanying drawing explanation
Fig. 1 shows that the first according to the method determining side airbag shape of the present invention is real
Execute the flow chart of mode；
Fig. 2 shows that the second according to the method determining side airbag shape of the present invention is real
Execute the flow chart of mode；
Fig. 3 shows the third reality according to the method determining side airbag shape of the present invention
Execute the flow chart of mode；
Fig. 4 shows a kind of original shape of side airbag and relatively optimizes the initial bit of coordinate system
Put.
Detailed description of the invention
Embodiments of the invention are described below in detail, and the example of described embodiment is shown in the drawings, its
In the most same or similar label represent same or similar element or there is same or like merit
The element of energy.The embodiment described below with reference to accompanying drawing is exemplary, is only used for explaining this
Bright, and be not construed as limiting the claims.
As it is shown in figure 1, the method for the determination side airbag shape of the present invention comprises the steps:
Step S1: select to optimize point, using all optimize point relatively optimize the position coordinates of coordinate system as
For determining the bladder shape variable of side airbag shape, owing to side airbag is by two panels
The monolithic gas cell fabric of identical left and right layout sews up synthesis, accordingly, it is determined that side airbag gas
Scrotiform shape monolithic to be determined gas cell fabric is in the shape under its flat presentation, so, selection
Optimize point also just it is believed that the point being on monolithic gas cell fabric profile, optimize coordinate system for determining
The two-dimensional coordinate system of flat shape；Assume to select 19 to optimize point, then this bladder shape variable is the most just
Including 19 position coordinateses optimizing point.
Step S2: obtain training bladder shape variable, and training bladder shape variable and air bag face
Amassing the corresponding relation between dummy's damage data, this dummy's damage data sets for side airbag
For meter, it is usually required mainly for examination dummy's rib deflection and dummy's abdominal part power index；Here, to training
With being interpreted as of bladder shape variable: known its with air bag area with dummy's damage data accurate corresponding
The bladder shape variable of relation, here, this training bladder shape variable can be from completed side
The data base of air bag shape design transfers, it is also possible to carrying out side airbag design every time
Time, by bladder shape variable being carried out less training sampling, more e.g. utilize air bag area meter
Calculate model and occupant injury is analyzed model and is calculated corresponding air bag area and dummy's damage data.
Step S31: utilize the corresponding relation between training bladder shape variable and air bag area, example
Function between reflection bladder shape variable and air bag area is obtained in this way by the method for interpolation or matching
The similar area model of relation, this Ke Rejin approximate model toolbar that such as may utilize MATLAB is real
Existing.
Step S32: utilize the corresponding relation between training bladder shape variable and dummy's damage data,
E.g. by the method for interpolation or matching obtain reflection bladder shape variable and dummy's damage data it
Between the approximation damage model of functional relationship, this such as may utilize the Ke Rejin approximate model work of MATLAB
Tool hurdle realizes, and includes dummy's rib deflection and the embodiment of dummy's abdominal part power for dummy's damage data,
This approximation damage model includes approximating thoracic injury model and approximation abdominal injury model.
Step S4: bladder shape variable is optimized sampling, it is thus achieved that optimize and use bladder shape variable.
Step S5: optimization bladder shape variable is separately input into similar area model and approximation damage
In model, to obtain the approximation air bag area corresponding with optimization bladder shape variable and approximation respectively
Dummy's damage data；Owing to optimization bladder shape variable is to utilize approximation air bag based on functional relationship
Area and approximation dummy's damage data calculate approximation air bag area and approximation dummy's damage data, this kind of meter
The arithmetic speed calculated quickly, therefore, can utilize phantom when being optimized sampling every time relatively
Calculating and select considerable sampled point, sampled point is the most, and optimum results is the most accurate, but the most correspondingly can
Reduce and optimize efficiency, here, be generally selected 5000 to 20000 sampled points, corresponding acquisition 5000
To 20000 groups of optimization bladder shape variablees, relatively satisfactory optimum results, wherein, gas can be obtained
As long as capsule configuration variable there being variable difference be i.e. considered different bladder shape variable.
Step S6: at the optimization gas making approximation dummy's damage data meet damage criterion constraints
The optimization bladder shape choosing the predetermined number minimum so that approximation air bag area in capsule configuration variable becomes
Amount solves as optimizing, and this predetermined number can set as required, but is difficult to excessive, and crossing conference increases
Amount of calculation, and it is unfavorable for optimizing the Fast Convergent solved, therefore, this predetermined number is for generally choosing 5
To 20 groups of optimization solutions.This damage criterion can be by obtaining, for dummy in China's new car value disciplines
Damage data includes dummy's rib deflection and the embodiment of dummy's abdominal part power, according to China's new car evaluation
Specification understands, and dummy's rib deflection is less than 22mm, and dummy's abdominal part power is less than 1KN.
Step S7: obtain and solve corresponding air bag area and dummy's damage data, with to excellent with optimizing
Air bag area and the dummy's damage data of dissolving correspondence confirm.
Step S8: choose making dummy's damage data meet in the optimization solution of damage criterion constraints
Make the optimization solution of air bag area minimum as target solution.
Step S9: determine side airbag shape according to target solution, this can be by target solution pair
Answer each optimize point carry out line and carry out line seamlessly transit realization.
In above-mentioned steps S2, for from the data base that completed side airbag shape designs
Transfer the embodiment of the training bladder shape variable matched, if this training bladder shape variable
Group number abundant, then through one suboptimization sampling be generally obtained with more accurate target solution；
And for when carrying out side airbag design every time, by being trained adopting to bladder shape variable
Sample, more e.g. utilize air bag areal calculation model and occupant injury analysis model to be calculated corresponding
Air bag area and the embodiment of dummy's damage data, if by the sampled point increasing training sampling
Mode improves similar area model and the precision of approximation damage model of acquisition, then on the one hand will increase
The amount of calculation of the inventive method, on the other hand, owing to the carrying out of training sampling is typically at each optimization point
The default excursion of position coordinates randomly selects, and therefore, even if increasing sampled point, also may not necessarily carry
High similar area model and approximation damage model precision in the spatial dimension being probably target solution, remove
Non-sampled point is very big, and otherwise the increase of sampled point has blindness.
In order to similar area model and approximation damage mould can be improved targetedly by less calculating
The precision of type so that the optimization solution selected is more and more accurate, until selecting target solution, the present invention
Method can also the optimization solution every time chosen be increased in training bladder shape variable, recycling
Training bladder shape variable through supplementing performs above-mentioned steps S31 and step S32, with pairing approximation
Area Model and approximation damage model are trained updating, to be iterated choosing optimization point.In approximation
After Area Model updates with approximation damage model, further wheel execution above-mentioned steps S4 is to step S7,
Optimize a little the most accurately to select.And perform above-mentioned steps S4 to the number of times of step S7, or
Claim the number of times that bladder shape variable is optimized sampling, can control by setting iteration stopping condition,
I.e. bladder shape variable is optimized employing, until meeting iteration stopping condition.For this kind
It is iterated choosing the embodiment optimizing point, above-mentioned steps S8 can meet making dummy's damage data
Optimizing in the optimization solution that sampling is chosen through last of damage criterion constraints, chooses so that air bag
The optimization solution of area minimum is as target solution.The above-mentioned embodiment being iterated choosing optimization point can be led to
Cross concrete steps as shown in Figure 2 to realize, the most after step s 7 increase step S7b:
Step S7b: judge whether to meet iteration stopping condition, the most then perform step S8, as otherwise
Perform step S10.
Step S10: increase to optimization solution, in training bladder shape variable, perform above-mentioned step afterwards
Rapid S31 and step S32.
May deposit to improve similar area model and approximation damage model the most targetedly
Precision in the spatial dimension of target solution, and search target solution that may be present at faster speed (i.e.
Accelerate to optimize the convergence rate solved), as it is shown on figure 3, the method for the present invention can also be in above-mentioned steps
Execution following steps after S7b:
Step S10a: carry out supplementary training sampling near solution optimizing, it is thus achieved that training is with supplementing air bag
Configuration variable, this neighbouring definition is: the excursion optimizing the position coordinates of point is reduced into and makes really
Each border of the fixed default excursion optimizing point relatively optimize solve the corresponding position coordinates optimizing point to
Scope determined by the distance of bias internal half, inward deflection distance in the most each border optimizes the excellent of solution correspondence
Change the distance of the half of some position coordinates in the corresponding direction.
Step S10b: obtain and train with supplementing the corresponding air bag area of bladder shape variable and dummy
Damage data.
Step S10: optimization is solved and training all increases to training air bag with supplementary bladder shape variable
In configuration variable, perform step S31 and step S32 afterwards, with to described similar area model and near
It is trained updating like damage model.
Above-mentioned iteration stopping condition is: the number of times that described bladder shape variable is optimized sampling reaches
The largest optimization sampling number preset, such as, set largest optimization sampling number as 200 to 1000 times,
Or the difference of the air bag area that the adjacent two suboptimization target solutions chosen of sampling are corresponding is less than or equal to presetting
Value, this preset value is e.g. less than or equal to 0.05.
Above-mentioned steps S1 selects optimize to put to farther include following steps:
Step S101: determine the original shape of side airbag shape and relatively optimize at the beginning of coordinate system
Beginning position, the setting of this original shape and initial position generally makes side airbag substantially covers
The whole region of driver's (or claiming dummy) thorax abdomen.
Step S102: select to optimize point on the profile of original shape, in order to set and optimize some position
The excursion of coordinate.
When determining air bag area and dummy's damage data of correspondence according to bladder shape variable, Ke Yisheng
Becoming the grid of the side airbag finite element analysis model corresponding with bladder shape variable, this is such as
The SFE CONCEPT Software Create of T-Solution company can have been passed through；Recycling grid obtains gas
Corresponding relation between capsule configuration variable and air bag area and dummy's damage data, this can facilitate logical
Cross to analyze in model grid input to existing air bag areal calculation model and occupant injury and obtain.Should
The method of kind is applicable to above-mentioned step S2, step S7 and step S10b.
Such as, for obtaining training bladder shape variable in step S2, and training bladder shape becomes
Measure and can farther include following steps with the corresponding relation between air bag area and dummy's damage data:
Step S201: bladder shape variable is trained sampling, becomes obtaining training bladder shape
Amount；
Step S202: generate the side airbag finite element corresponding with training bladder shape variable
Analyze the grid of model, as training grid；
Utilize training grid, such as, training grid is inputted to air bag areal calculation model and occupant injury
Analyze in model, obtain training with between bladder shape variable and air bag area and dummy's damage data
Corresponding relation.
The most such as, corresponding air bag area is solved and dummy damages number for step S7 obtains with optimizing
According to farther including following steps:
Step S701: generate and optimize the net solving corresponding side airbag finite element analysis model
Lattice, as optimizing grid.
Step S702: utilize and optimize grid, such as, will optimize grid and input to air bag areal calculation model
Analyze in model with occupant injury, obtain and solve corresponding air bag area and dummy's damage data with optimizing.
Above-mentioned steps S201 is trained sampling to bladder shape variable and comprises the steps that employing EXPERIMENTAL DESIGN
Method is trained sampling to described bladder shape variable.To bladder shape variable in above-mentioned steps SS4
It is optimized sampling and comprises the steps that employing test design method is optimized sampling to bladder shape variable.
Above-mentioned steps S10a carries out near optimization solves supplementary training sampling and comprises the steps that employing EXPERIMENTAL DESIGN
Method carries out supplementary training sampling near optimization solves.Here, use test design method to adopt
Restrain quickly with making to optimize solution relative to the mode randomly selected, reduce the number of times needing iteration,
This test design method e.g. optimal Latin hypercube test design method, orthogonal experiment design method, all
Even test design method etc..
In order to highlight the present invention can be relatively small amount of calculation obtain more accurate optimum results spy
Point, the group of above-mentioned optimization bladder shape variable bladder shape variable being optimized sampling and obtain
Number (being i.e. optimized the sampled point of sampling) is at least and described bladder shape variable is trained sampling
And obtain 10 times of the group number (being i.e. trained the sampled point of sampling) of training bladder shape variable,
Especially for 30 times to 100 times.
For the ease of setting the excursion of each position coordinates optimizing point, as shown in Figure 4, above-mentioned excellent
Change the first coordinate axes a direction extension along backrest of coordinate system, side the most upwards can be selected
To for forward, optimize the second coordinate axes b of coordinate system at the perpendicular extended along length of wagon direction
Upper vertical with the first coordinate axes a, can select basic forward direction is forward.And optimize coordinate system
Initial point o can be chosen as dummy H point so that optimize point position coordinates be relative dummy H point
Position coordinates, so can avoid the method for the determination side airbag of the present invention to be become by vehicle size
The impact changed.
Owing to side airbag is side airbag gas on the limit that the second coordinate axes b is located behind forward
The installation limit of capsule, therefore, optimizes point and is predominantly on being connected with installing limit at two ends of side airbag
On the contour line connect.The excursion optimizing some position coordinates can refer to following manner setting, can make
The position coordinates optimizing point is sat on correspondence position edge relative to original shape at the possible side airbag that makes
Highest distance position that parameter direction is overhanging and possible make side airbag relative to original shape in corresponding position
Put and change between the deepest position that change in coordinate axis direction inside contracts.
In order to simplify design further, can make side airbag with being positioned at of being connected of limit is installed
Above and below limit be respectively straightway, this make optimize point be only included in the second coordinate axes b forward
On upper located anteriorly limit, and it is positioned on four angle points of original shape.On this basis, Fig. 4 gives
Go out a kind of original shape and be chosen as square embodiment, the above and lower section of this original shape
Limit is parallel with the second coordinate axes b, located anteriorly parallel with the first coordinate axes a with the limit at rear, selects
The point that optimizes include square four angle point P1, P17, P18 and P19, and the second coordinate axes b's
The most located anteriorly optimization point P2 to P16 optimized on limit, here, between usual on this optimization limit
Choosing optimization point every 20～30mm, in the embodiment shown in Fig. 4, original shape is at the first coordinate axes
A length of 480mm on direction, therefore, by be spaced 30mm choose optimization point in the way of, i.e. excellent
Change and 17 optimization points are set on limit altogether；And be side on the limit being located behind forward of the second coordinate axes b
The installation limit of face air bag, therefore, in addition to putting angle point P18 and P19 as optimization, it is not necessary to volume
Outer arrange other optimize point, so, one group of bladder shape variable i.e. includes the position coordinates of P1 to P19
P1 (a, b) to P19 (a, b).
When setting the excursion of each position coordinates optimizing point, refer in the following manner and carry out:
When determining original shape and initial position, if substantially to cover driver's (or claiming dummy) chest and abdomen
The whole region in portion is for requiring, then for four angle points P1, P17, P18 and P19 at the first coordinate axes
Initial position on a direction is essentially the maximum to dependent variable, inwardly translates e.g. 30mm's
Certain distance is as the minima to dependent variable, and this certain distance is typically smaller than equal at the first coordinate axes
On a direction, (the most above-mentioned choosing on optimization limit optimizes what point was used with the adjacent distance optimized between point
Interval), for the embodiment shown in Fig. 4, optimize some P1, P19 negative sense along the first coordinate axes a
Mobile 30mm, optimizes some P17, P18 point and moves 30mm along the first coordinate axes a forward；For four
Individual angle point excursion on the second coordinate axes b direction, for optimizing optimization point P1, the P17 on limit
The certain distance of e.g. 100mm is moved as the maximum to dependent variable along the second coordinate axes b forward
Value, moves along the second coordinate axes b negative sense until the position of dummy's rib deformation amount sensor line is right
The minima of dependent variable, here, the direction along the first coordinate axes a is provided with upper, middle and lower on seat
Three dummy's rib deformation amount sensors, and be positioned at optimization point P18, the P19 installed on limit and sit second
Position coordinates P18 (b) on parameter b direction, P19 (b) are fixed value, it is not necessary to optimize.Right
Other beyond non-angle point optimizes some P2 to P16, only need to optimize along the second coordinate axes b direction
Position coordinates P2 (b) is to P16 (b), and these optimizations o'clock are moved such as along the second coordinate axes b forward
Be the certain distance of 100mm as corresponding variable maximum, move directly along the second coordinate axes b negative sense
Position to dummy's rib deformation amount sensor line is the minima to dependent variable.
The concrete application example of one given below: use original shape as shown in Figure 4, optimization point to set
Putting, the excursion optimizing some position coordinates uses above-mentioned set-up mode；Set above-mentioned iteration stopping bar
Largest optimization sampling number in part is 500 times, and preset value is 0.001；Use latin square experiment method
Carry out various sampling；The sampled point being set for training sampling is 200 points, i.e. obtains 200 groups of gas
Capsule configuration variable；Set the sampled point being every time optimized sampling as 10000 points, set choosing every time
Take 10 and optimize solution, set the sampled point every time carrying out supplementary training sampling as 50 points.
Optimum results is: eventually passes through 420 suboptimization samplings and terminates, obtains the side after an optimization
Air bag shape, the side airbag shape that application obtains, the area of side airbag is slightly
Increasing, area increase is concentrated mainly on epipleural bone and abdominal part protection zone, driver (dummy) rib
Deflection and abdominal part power are respectively 20.9mm and 0.91kN, have obtained changing significantly compared with before optimization
Kind.There is certain discontinuity in view of the shape after optimizing, enter according to the bladder shape after optimizing
Having gone the transition processing of local, used the installing side airbag shape after transition processing, driver's rib deforms
Amount does not changes compared with before transition processing substantially with abdominal part power, it is thus achieved that identical protected effect.
The structure of the present invention, feature and effect effect are described in detail above according to graphic shown embodiment
Really, the foregoing is only presently preferred embodiments of the present invention, but the present invention does not implements to limit shown in drawing
Scope, every change made according to the conception of the present invention, or it is revised as the equivalence enforcement of equivalent variations
Example, still without departing from description with diagram contained spiritual time, all should be within the scope of the present invention.
Claims (9)
1. the method determining side airbag shape, it is characterised in that including:
Select to optimize point, comprise determining that the original shape of side airbag shape and relatively optimize seat
The initial position of mark system, selects to optimize point on the profile of described original shape；
All points that optimize relatively are optimized the position coordinates of coordinate system determine side airbag as being used for
The bladder shape variable of shape；
Obtain training bladder shape variable, and bladder shape variable and air bag area and dummy are used in training
Corresponding relation between damage data；
Utilize the corresponding relation between training bladder shape variable and air bag area, obtain reflection air bag
The similar area model of functional relationship between configuration variable and air bag area；
Utilize the corresponding relation between training bladder shape variable and dummy's damage data, it is thus achieved that reflection
The approximation damage model of functional relationship between bladder shape variable and dummy's damage data；
Described bladder shape variable is optimized sampling, it is thus achieved that optimize and use bladder shape variable；
Optimization bladder shape variable is separately input into described similar area model and approximation damage model
In, to obtain the approximation air bag area corresponding with optimization bladder shape variable and approximation dummy respectively
Damage data；
Become in the optimization bladder shape making approximation dummy's damage data meet damage criterion constraints
Amount is chosen the optimization bladder shape variable of the predetermined number minimum so that approximation air bag area as excellent
Dissolve；
Obtain and solve corresponding air bag area and dummy's damage data with optimizing；
Choose so that air bag making dummy's damage data meet in the optimization solution of damage criterion constraints
The optimization solution of area minimum is as target solution；
Side airbag shape is determined according to described target solution.
Method the most according to claim 1, it is characterised in that described method also includes:
Described optimization solution is increased in described training bladder shape variable, and to described similar area
Model and approximation damage model are trained updating；
Described described bladder shape variable be optimized sampling include:
Described bladder shape variable is optimized sampling, until meeting iteration stopping condition；
Described choose in the optimization solution making dummy's damage data meet damage criterion constraints so that
The optimization solution of air bag area minimum includes as target solution:
Selecting through last optimization sampling of damage criterion constraints is met making dummy's damage data
In the optimization solution taken, choose the optimization solution so that air bag area minimum as target solution.
Method the most according to claim 2, it is characterised in that described method also includes:
Supplementary training sampling is carried out near solution, it is thus achieved that training becomes with supplementing bladder shape described optimization
Amount；
Obtain with training with supplementing the corresponding air bag area of bladder shape variable and dummy's damage data；
Described optimization is solved and training increases to described training bladder shape with supplementary bladder shape variable
In variable, and it is trained updating to described similar area model and approximation damage model.
The most according to the method in claim 2 or 3, it is characterised in that described iteration stopping condition
For: the number of times that described bladder shape variable is optimized sampling reaches default largest optimization sampling time
Number, or the difference of air bag area corresponding to the adjacent two suboptimization target solutions chosen of sampling is less than or equal to
Preset value.
5. according to the method described in claim 1,2 or 3, it is characterised in that described acquisition is trained
Use bladder shape variable, and training is with between bladder shape variable and air bag area and dummy's damage data
Corresponding relation include:
Described bladder shape variable is trained sampling, to obtain training bladder shape variable；
Generate the side airbag finite element analysis mould corresponding with described training bladder shape variable
The grid of type, as training grid；
Utilize described training grid, obtain training bladder shape variable and air bag area and dummy's damage
Corresponding relation between data；And/or,
Air bag area and dummy's damage data that described acquisition is corresponding with optimizing solution include:
Generate and the described grid optimizing the corresponding side airbag finite element analysis model of solution, make
For optimizing grid；
Utilize described optimization grid, obtain and solve corresponding air bag area and dummy's damage data with optimizing.
Method the most according to claim 5, it is characterised in that described described bladder shape is become
Amount is trained sampling and includes: use test design method to be trained adopting to described bladder shape variable
Sample；And/or,
Described described bladder shape variable is optimized sampling includes: use test design method to institute
State bladder shape variable and be optimized sampling.
Method the most according to claim 5, it is characterised in that described bladder shape variable is entered
Described bladder shape is at least become by the group number of the optimization bladder shape variable that row optimizes sampling and obtains
Amount is trained sampling and obtains 10 times of the group number of training bladder shape variable.
Method the most according to claim 5, it is characterised in that described utilize described training grid,
The corresponding relation obtained between training bladder shape variable and air bag area and dummy's damage data includes:
Described training grid is separately input into air bag areal calculation model and occupant injury is analyzed in model,
To obtain the air bag area corresponding with training bladder shape variable and dummy's damage data respectively；And/
Or,
Described utilize described optimization grid, obtain and solve corresponding air bag area and dummy's damage with optimizing
Data include:
Described optimization grid is separately input into air bag areal calculation model and occupant injury is analyzed in model,
Corresponding air bag area and dummy's damage data is solved with optimizing to obtain respectively.
9. according to the method described in claim 1,2 or 3, it is characterised in that described optimization coordinate
System is two-dimensional coordinate system, and the first coordinate axes of described optimization coordinate system extends along the direction of backrest,
Second coordinate axes of described optimization coordinate system on the perpendicular extended along length of wagon direction with described
First coordinate axes is vertical.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410242898.3A CN103984838B (en) | 2014-06-03 | 2014-06-03 | A kind of method determining side airbag shape |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410242898.3A CN103984838B (en) | 2014-06-03 | 2014-06-03 | A kind of method determining side airbag shape |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103984838A CN103984838A (en) | 2014-08-13 |
| CN103984838B true CN103984838B (en) | 2016-10-05 |
Family
ID=51276807
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410242898.3A Active CN103984838B (en) | 2014-06-03 | 2014-06-03 | A kind of method determining side airbag shape |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103984838B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104527562B (en) * | 2014-12-19 | 2018-01-19 | 浙江吉利汽车研究院有限公司 | A kind of vehicle restraint systems design and develop method |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101013448A (en) * | 2006-02-02 | 2007-08-08 | 福特全球技术公司 | Vehicular supplemental restraint device simulation using finite element modeling |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030198767A1 (en) * | 2002-04-19 | 2003-10-23 | Breed David S. | Airbags including barrier coating and method of manufacture of same |
| WO2002086670A2 (en) * | 2001-04-24 | 2002-10-31 | Breed Automotive Technology, Inc. | Simplified modeling software interface and method |
-
2014
- 2014-06-03 CN CN201410242898.3A patent/CN103984838B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101013448A (en) * | 2006-02-02 | 2007-08-08 | 福特全球技术公司 | Vehicular supplemental restraint device simulation using finite element modeling |
Non-Patent Citations (3)
| Title |
|---|
| LY轿车侧面安全气囊设计与开发;付艳;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20140415(第04期);第C034-164页 * |
| 基于零部件试验的汽车侧面碰撞假人伤害的研究;董学勤等;《汽车工程》;20121025;第34卷(第10期);第894-897页 * |
| 汽车侧面碰撞结构改进和侧面头胸部安全气囊的优化研究;吴亚军;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20100115(第01期);第C035-33页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103984838A (en) | 2014-08-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tuncali et al. | Utilizing S-TaLiRo as an automatic test generation framework for autonomous vehicles | |
| CN108121205B (en) | Path planning method, system and medium for multiple parking scenes | |
| CN103909522B (en) | A kind of Six-DOF industrial robot is by the method in unusual territory | |
| CN107256289A (en) | The method for building up of car crass reduced parameter FEM model | |
| CN105335797A (en) | Automatic parking locus optimization method based on full-simultaneous dynamic optimization framework | |
| CN109885883A (en) | A kind of control method of the unmanned vehicle transverse movement based on GK clustering algorithm model prediction | |
| CN109991974A (en) | Automatic Pilot path follower method, device and control equipment | |
| CN109344498B (en) | Design implementation method of wedge module | |
| CN103186689B (en) | Electromagnetic-field simulation analytical approach | |
| CN103984838B (en) | A kind of method determining side airbag shape | |
| CN104573201A (en) | Quality matching design method of precision machine tool | |
| Gao et al. | Accurate pseudospectral optimization of nonlinear model predictive control for high-performance motion planning | |
| CN104345687B (en) | A kind of tool sweep volume modeling method | |
| CN103366056A (en) | Machine part cylindricity error measuring method based on finite element analysis | |
| CN104573246A (en) | Front impact conceptual model design method for front structure of automobile | |
| CN104282033A (en) | Integrated lane changing behavior simulation method oriented towards vehicle animation simulation | |
| CN109063409A (en) | A kind of light-weight design method and apparatus of vehicle | |
| CN105335703A (en) | Traffic scene abnormity detection method based on movement reconstruction technology | |
| Ye et al. | A study of the novel vision guided IV trajectory tracking control system based on expected yaw velocity | |
| Han et al. | Adjoint method for aerodynamic shape improvement in comparison with surface pressure gradient method | |
| CN106004880A (en) | Method and system for determining primary direction of force resulting from vehicle collision | |
| CN103017722A (en) | CAE (Computer Aided Engineering)-based method for measuring vehicle body deformation by simulating head-on collision | |
| Brembeck et al. | Real-time capable path planning for energy management systems in future vehicle architectures | |
| CN105835990A (en) | Tool and method for positioning H point of total arrangement space verification model of passenger car | |
| CN104881531B (en) | A kind of Autoform punch formings information is to the mapping method of collision simulation model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 230601 Anhui Province, Hefei City Industrial Park, the Peach Blossom Road No. 669 Patentee after: Anhui Jianghuai Automobile Group Limited by Share Ltd Address before: 230601 Anhui Province, Hefei City Industrial Park, the Peach Blossom Road No. 669 Patentee before: Anhui Jianghuai Automotive Co., Ltd. |