CN104834772A - Artificial-neural-network-based inverse design method for aircraft airfoils/wings - Google Patents
Artificial-neural-network-based inverse design method for aircraft airfoils/wings Download PDFInfo
- Publication number
- CN104834772A CN104834772A CN201510198979.2A CN201510198979A CN104834772A CN 104834772 A CN104834772 A CN 104834772A CN 201510198979 A CN201510198979 A CN 201510198979A CN 104834772 A CN104834772 A CN 104834772A
- Authority
- CN
- China
- Prior art keywords
- wing
- coefficient
- aerofoil
- design
- neural network
- 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.)
- Granted
Links
- 238000013461 design Methods 0.000 title claims abstract description 69
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 44
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 13
- 238000005516 engineering process Methods 0.000 claims abstract description 10
- 238000012549 training Methods 0.000 claims description 13
- 230000035939 shock Effects 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 2
- FNYLWPVRPXGIIP-UHFFFAOYSA-N Triamterene Chemical compound NC1=NC2=NC(N)=NC(N)=C2N=C1C1=CC=CC=C1 FNYLWPVRPXGIIP-UHFFFAOYSA-N 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000007429 general method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Landscapes
- Other Investigation Or Analysis Of Materials By Electrical Means (AREA)
- Structures Of Non-Positive Displacement Pumps (AREA)
Abstract
The invention belongs to the technical field of design of aircrafts and in particular relates to an artificial-neural-network-based inverse design method for aircraft airfoils/wings. The method comprises the following steps: rebuilding expression modes of airfoils/wings by using PARSEC parameterization method of airfoils/wings, and implementing inverse design technology by using an artificial neural network (ANN) algorithm. By virtue of the artificial-neural-network-based inverse design method for aircraft airfoils/wings, a conventional cut-and-try method with complex design and low efficiency for aircraft airfoils/wings is put aside; a relation between the aerodynamic performance of airfoils/wings and the geometric profile of aircraft airfoils/wings is directly built; the artificial-neural-network-based parameterization inverse design for aircraft airfoils/wings is implemented. The artificial-neural-network-based inverse design method for aircraft airfoils/wings is high in speed and is highly suitable for the overall design of aircrafts, especially the initial design; the artificial neural network algorithm is in place, so that the generated results are highly accurate.
Description
Technical field
The invention belongs to airplane design technical field, be specifically related to a kind of aircraft wing/wing inverse design method.
Background technology
Mimetic design method in aircraft components design, i.e. a kind of " required i.e. gained ", directly can obtain the geometric data of satisfactory aircraft components according to given aerodynamic performance requirements.At present, the method of common aircraft wing/wing design be traditional, loaded down with trivial details and poor efficiency enumerate-alternative manner (cut-and-try), namely by calculating the aeroperformance of the design result of aircraft wing/wing, fed back to and the change made new advances is made to geometric data (often such change and the relevance of aeroperformance are not very strong, and present larger randomness), then carry out new aeroperformance to calculate, so iterate until obtain the new aerofoil/wing meeting required aeroperformance.Due to the cause that such method for designing specific aim is not strong, the time quantum of consumption is comparatively large, the job requirement in be unfavorable for airplane design work especially initial design stage.
About the mimetic design method of the aerofoil profile/wing of aircraft, object directly forms contacting of aircraft wing/wing aerodynamic performance and geometric data, realize direct correlation from aeroperformance to geometric data, thus obtain more fast, more direct aerofoil profile/Wing design method.At present, the technology in aircraft components mimetic design field usually utilizes various intelligent algorithm, carries out the mapping of aeroperformance data and geometry data.But being limited to aircraft components design the complex nature of the problem, the intelligent algorithm taked is not quite similar, and the effect obtained is also unsatisfactory.Based on this present situation, the present invention proposes the aircraft wing/wing inverse design method based on artificial neural network.
Summary of the invention
In order to overcome above-mentioned the deficiencies in the prior art, the present invention proposes a kind of aircraft wing based on artificial neural network/wing inverse design method.
Aircraft wing based on artificial neural network provided by the invention/wing inverse design method, is mainly divided into 5 steps:
(1) first, utilize aerofoil profile/wing PARSEC (Sobieczky H. Parametric airfoils and wings [M] //Recent Development of Aerodynamic Design Methodologies. Vieweg+ Teubner Verlag, 1999:71-87.) expression way of parametric method reconstruct aerofoil profile, namely uses 11 PARSEC formal parameter (leading-edge radius r
le, up/down aerofoil maximum gauge X
upand X
lo, up/down aerofoil maximum gauge correspondence position Z
upand Z
lo, up/down aerofoil vertex curvature Z
xxupand Z
xxlo, trailing edge width △ Z
tE, trailing edge vertical height Z
tE, trailing edge angle of wedge β
tE, trailing edge direction angle alpha
tE) simulate airfoil geometry situation, the aeroperformance of aerofoil profile is summarized with 6 Aerodynamics (lift coefficient CL, resistance coefficient CD, moment coefficient CM, cruise efficiency MCL/CD, coefficient of frictional resistance CDF, drag due to shock wave coefficient CDW).
Top airfoil and lower aerofoil parameter of curve expression formula are for shown in formula (1):
（1）
A
nfor multinomial coefficient, for top airfoil, coefficient a
nprovided by matrix equation (2):
（2）
For lower aerofoil, coefficient a
nprovided by matrix equation (3), and top airfoil is similar, the parameter being about to characterize top airfoil configuration characteristic amount changes the corresponding parameter of lower aerofoil into:
（3）
Obtain fitting coefficient, just can set up contacting of geometric parameter and actual aerofoil profiles.
On the basis of PARSEC geometric parameter method, be optimization object to aerofoil profile/wing inverse design with geometric parameter.
(2) then, moulded the wing section of different length position by the selection of aerofoil profile and size scaling by the geometric parameter data of aerofoil profile; In conjunction with wing relative thickness parameter and the torsion angle parameter (general 6 can reach anticipate accuracy) of different length position, utilize spline curve fitting wing wing, obtain the geometric data of wing, implementing procedure is see Fig. 3.The parameter that corresponding wing aerodynamic performance describes comprises: cruise efficiency MCL/CD, airfoil lift coefficient CLWING, wing drag coefficient CDWING, pressure drag coefficient CDP, induced drag coefficient CDI, drag due to shock wave coefficient CDW and wing moment coefficient CMQING.
(3) then, according to the geometric data result about aerofoil profile and wing that preceding step obtains, be applied on a large amount of aerofoil profiles, to form airfoil geometry database.The success or not of this technology depends on that can geometric database be accomplished: 1. aerodynamic data and geometric data cover enough large scope, make this mimetic design technology to have use value; 2. whether the data of database under a certain classification are enough substantial, to guarantee that the study of the artificial neural network required for (4) step next can be not only complete but also accurate.At present, number of training object determines do not have general method, it is generally acknowledged, sample is very few may make the expression of network abundant not, thus causes the extrapolability of network inadequate, sample too much there will be sample redundancy phenomena, both added network training burden, and may occur that again quantity of information is superfluous, make network occur Expired Drugs, database, through the training of artificial neural network, defines available mimetic design artificial neural network.User now arranges required mimetic design Aerodynamic at input end again, then mimetic design artificial neural network now can export the wing/airfoil geometry parameter that meet this performance automatically.User can obtain the profile of wing/aerofoil profile according to this result, for design effort further.
Applying before the database that arranged carries out artificial neural network training, the classification that needs carry out for its geometric data, to guarantee that the aerofoil profile data as the training input end of mimetic design artificial neural network can geometrically be close as far as possible.This way is based on considering on a mechanics: the aeroperformance that close geometric shape is formed should be close.Such understanding, contributes to the training difficulty lowering artificial neural network.Sorting technique adopts SOM algorithm (Kohonen T, Hynninen J, Kangas J. et al. Som pak:The self-organizing map program package [J]. Report A31, Helsinki University of Technology, Laboratory of Computer and Information Science, 1996.).
(4) then on this basis, by artificial neural network (ANN) algorithm (Zeidenberg M. Neural networks in artificial intelligence [M]. Ellis Horwood, 1990.), in order to realize mimetic design, carry out the training of relevant people artificial neural networks.About the training method of neural network, adopt GRNN algorithm (Specht D F. A general regression neural network. Neural Networks [J], IEEE Transactions on, 1991. 2 (6): p. 568-576.) (through comparing, its target related coefficient level and network generalization are all better than BP algorithm and RBF algorithm): input end is that the aeroperformance of wing/aerofoil profile (can be accepted or rejected according to the demand of mimetic design, such as can choose lift-drag ratio coefficient during aircraft initial designs), output terminal is the geometric data of corresponding wing/aerofoil profile.
(5) after mimetic design artificial neural network has been trained, just can train at this work carrying out mimetic design on complete artificial neural network, i.e. aerofoil profile/wing aerodynamic the performance parameter of user required by the input end input of network, now the result of output terminal is exactly the geometric parameter of the aerofoil profile of Aerodynamic required by required correspondence.By the aerofoil method of reducing (i.e. formula (1)) in step (1), just final result can be obtained.
Compared with prior art, what the present invention bypassed the loaded down with trivial details and poor efficiency of traditional aerofoil profile/wing design by mimetic design enumerates-alternative manner (cut-and-try), but directly sets up the relation of aerofoil profile/wing aerodynamic performance and aerofoil profile/wing geometric shape.Simultaneously it and general database search are different: it can utilize the intellectual technology height " interpolation " of artificial neural network obtain required for result.Feature of the present invention: be on the one hand quick, among the process of the overall design being extremely suitable for aircraft especially initial designs; On the other hand because using artificial neural networks algorithm puts in place, make the result of generation very accurate.This technology obtains inspection in numerical simulation: be Mach number 0.75, the angle of attack 2.53 °, Reynolds number 23 for operating mode, 000,000, lift-drag ratio is 25.8, lift coefficient is 0.5, type resistance coefficient is 0.0105, induced drag coefficient is 0.009, and drag due to shock wave coefficient is 0.00017, and moment coefficient is 0.23; Airfoil lift coefficient is 0.43, resistance coefficient be 0.0095 and moment coefficient be-0.13 wing profile supercritical wing design, this technology applies the wing obtained and the relative error of expecting on aerodynamic parameter all within 2%, result satisfactory (referring to hereinafter " embodiment " part).
Accompanying drawing explanation
Fig. 1 is aerofoil profile/wing inverse design concept map.
Fig. 2 is aerofoil profile/wing inverse design GRNN network mode used figure.
Fig. 3 is the conversion process figure of aerofoil profile/wing inverse design aerofoil profile parameter to wing geometric parameter, and it corresponds to step 2.
Fig. 4 is wing PARSEC parameter manifestation mode in the present invention.
Fig. 5 is that in object lesson, mimetic design obtains aerofoil profiles.
Fig. 6 is wing section geometric configuration.
Fig. 7 is relative thickness distribution (left side) and torsion angle distribution (right side) in prediction wing section.
Fig. 8 is the distribution of wing isobar.
Fig. 9 is the pressure distribution in 6 wing sections.
Embodiment
Embodiment 1:
Designing requirement: under a cruise Mach number Ma=0.785, design point is the aerofoil profile of following parameter:
Design Mach number Ma=0.71
Design angle of attack á=2.53
Reynolds number Re=23,000,000
First airfoil database is set up.Textural in profile, aerofoil profile is relatively simple.208 reconstruct aerofoil profiles are comprised in airfoil database in this example.Each aerofoil profile in database will have good aeroperformance; And the geometric configuration of aerofoil profile will have span large as far as possible, namely obvious shape differentiation to be had between aerofoil profile.The high efficiency of such guarantee airfoil database and integrality.The span that table 1 and table 2 are respectively aerofoil profile moving parameter and formal parameter gathers.
Table 1 aerofoil profile aerodynamic parameter gathers
Table 2 aerofoil profiles parameter gathers
| Minimum Value | Maximum Value | |
| r le | 0.0034 | 0.0306 |
| X up | 0.2927 | 0.5074 |
| X lo | 0.275 | 0.4577 |
| Z up | 0.0301 | 0.1116 |
| ｜Z lo｜ | 0.0186 | 0.0923 |
| ｜Z xxup｜ | 0.02 | 1.1086 |
| Z xxlo | 0.096 | 0.919 |
| △Z TE | 0 | 0.0163 |
| Z TE | -0.0419 | 0.0082 |
| β TE | 0.01 | 8.4622 |
| α TE | 2.1599 | 16.6992 |
Apply the artificial neural network (input and output mode is as shown in Figure 1) based on mimetic design, carry out training and input and the aeroperformance desired by this example, obtaining the aerofoil profile result of mimetic design, as shown in table 3 and Fig. 5:
Table 3 Airfoil Inverse Design result
。
After obtaining mimetic design aerofoil profiles, Flow Field Calculation is carried out to it, the calculating design aerodynamic parameter of gained and the expection aerodynamic parameter of target call are contrasted.The relative error of major design aerodynamic parameter is all within allowed band.
The error analysis of table 4 Airfoil Inverse Design
。
Embodiment 2:
Designing requirement: designing lift-drag ratio is 25.8, and lift coefficient is 0.5, type resistance coefficient is 0.0105, and induced drag coefficient is 0.009, and drag due to shock wave coefficient is 0.00017, and moment coefficient is 0.23; Airfoil lift coefficient is 0.43, resistance coefficient be 0.0095 and moment coefficient be-0.13 wing profile.
Wherein main aerodynamic parameter (wing lift-drag ratio and lift coefficient) is proposed by deviser, and remaining secondary aerodynamic parameter interpolation method provides, as table 5 with reference to group pneumatic input range of 214 in main aerodynamic parameter and database.
Table 5 wing aerodynamic parameter gathers
。
First by SOM neural network, the wing data in database are classified according to aeroperformance, take out one group of data corresponding to the pneumatic input of target, select GRNN neural network to calculate.
To expect according to wing by mimetic design method and calculate 78 wing formal parameters by aeroperformance, construct wing profile according to parametric method and wing construction method formal parameter.Wherein, the geometric configuration in 6 wing sections respectively as shown in Figure 6; As shown in Figure 7, corresponding aeroperformance as shown in Figure 9 for the relative thickness distribution in each wing section and torsion angle distribution; The form of whole wing and aeroperformance are as shown in Figure 8.
The calculating design aerodynamic parameter of gained and the expection aerodynamic parameter of target call are contrasted.The relative error of efficiency factor is 0.39%, and the relative error of multi-wall interference lift coefficient is 2.22%, and the error of airfoil lift coefficient is 1.33%.The relative error of three main aerodynamic parameters is all within allowed band.
Table 6 wing formal parameter predicated error is analyzed
。
Claims (1)
1., based on aircraft wing/wing inverse design method of artificial neural network, it is characterized in that concrete steps are:
(1) first, utilize the expression way of aerofoil profile/wing PARSEC parametric method reconstruct aerofoil profile, namely use 11 PARSEC formal parameters: leading-edge radius r
le, up/down aerofoil maximum gauge X
upand X
lo, up/down aerofoil maximum gauge correspondence position Z
upand Z
lo, up/down aerofoil vertex curvature Z
xxupand Z
xxlo, trailing edge width △ Z
tE, trailing edge vertical height Z
tE, trailing edge angle of wedge β
tE, trailing edge direction angle alpha
tEsimulate airfoil geometry situation, with 6 Aerodynamics: lift coefficient CL, resistance coefficient CD, moment coefficient CM, cruise efficiency MCL/CD, coefficient of frictional resistance CDF, drag due to shock wave coefficient CDW describe the aeroperformance of aerofoil profile; So top airfoil and lower aerofoil parameter of curve expression formula are for shown in formula (1):
（1）
A
nfor multinomial coefficient, for top airfoil, coefficient a
nprovided by matrix equation (2):
（2）
For lower aerofoil, coefficient a
nprovided by matrix equation (3):
（3）
Obtain fitting coefficient, just set up contacting of geometric parameter and actual aerofoil profiles;
On the basis of PARSEC geometric parameter method, be optimization object to aerofoil profile/wing inverse design with geometric parameter;
(2) the geometric parameter data of aerofoil profile then, are utilized to mould the wing section of different length position by the selection of aerofoil profile and size scaling; In conjunction with wing relative thickness parameter and the torsion angle parameter of different length position, utilize spline curve fitting wing wing, obtain the geometric data of wing; The parameter that corresponding wing aerodynamic performance describes comprises: cruise efficiency MCL/CD, airfoil lift coefficient CLWING, wing drag coefficient CDWING, pressure drag coefficient CDP, induced drag coefficient CDI, drag due to shock wave coefficient CDW and wing moment coefficient CMQING;
(3) then, according to the result of the geometric data about aerofoil profile and wing that preceding step obtains, be applied on a large amount of aerofoil profiles, to form airfoil geometry database; Requirement is accomplished: (A) aerodynamic data and geometric data cover enough large scope, make this mimetic design technology to have use value; (B) data of database under a certain classification are enough substantial, to guarantee that the study of the artificial neural network required for later step (4) can be not only complete but also accurate; Database, through the training of artificial neural network, forms available mimetic design artificial neural network;
Adopt SOM algorithm, airfoil geometry data are classified, to guarantee that the aerofoil profile data as the training input end of mimetic design artificial neural network can geometrically be close as far as possible;
(4) on this basis, by artificial neural network (ANN) algorithm, the training of mimetic design artificial neural network is carried out; Wherein, about the training of neural network, adopt GRNN algorithm: input end is the aeroperformance data of wing/aerofoil profile, and these aeroperformance data can be accepted or rejected according to the demand of mimetic design, output terminal is the geometric data of corresponding wing/aerofoil profile;
(5) after mimetic design artificial neural network has been trained, the work carrying out mimetic design on complete artificial neural network is trained at this, i.e. aerofoil profile/wing aerodynamic the performance parameter of user required by the input end input of network, now the result of output terminal is exactly the geometric parameter of the aerofoil profile of Aerodynamic required by required correspondence; By the aerofoil method of reducing in (1) step, obtain final result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510198979.2A CN104834772B (en) | 2015-04-22 | 2015-04-22 | Aircraft wing based on artificial neural network/wing inverse design method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510198979.2A CN104834772B (en) | 2015-04-22 | 2015-04-22 | Aircraft wing based on artificial neural network/wing inverse design method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104834772A true CN104834772A (en) | 2015-08-12 |
| CN104834772B CN104834772B (en) | 2018-01-12 |
Family
ID=53812656
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510198979.2A Expired - Fee Related CN104834772B (en) | 2015-04-22 | 2015-04-22 | Aircraft wing based on artificial neural network/wing inverse design method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104834772B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107014451A (en) * | 2017-05-03 | 2017-08-04 | 东南大学 | The method of ultrasonic flow sensor coefficient is speculated based on generalized regression nerve networks |
| CN108733914A (en) * | 2018-05-17 | 2018-11-02 | 复旦大学 | Transonic airfoil Natural Laminar Flow delay based on artificial neural network turns to twist design method |
| CN110084511A (en) * | 2019-04-25 | 2019-08-02 | 广东工业大学 | A kind of unmanned plane matching method, device, equipment and readable storage medium storing program for executing |
| CN110321588A (en) * | 2019-05-10 | 2019-10-11 | 中车青岛四方车辆研究所有限公司 | Rail vehicle aerodynamic Drag Calculation method based on numerical simulation |
| DE102018210894A1 (en) * | 2018-07-03 | 2020-01-09 | Siemens Aktiengesellschaft | Design and manufacture of a turbomachine blade |
| WO2020112023A1 (en) * | 2018-11-26 | 2020-06-04 | Agency For Science, Technology And Research | Method and system for predicting performance in electronic design based on machine learning |
| CN112784508A (en) * | 2021-02-12 | 2021-05-11 | 西北工业大学 | Deep learning-based airfoil flow field rapid prediction method |
| CN113849910A (en) * | 2021-09-23 | 2021-12-28 | 四川大学 | Dropout-based BiLSTM network wing resistance coefficient prediction method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101458735A (en) * | 2008-12-31 | 2009-06-17 | 重庆大学 | Aerofoil with high lift-drag ratio |
| US20090157364A1 (en) * | 2007-12-18 | 2009-06-18 | Airbus Espana, S.L.. | Method and system for a quick calculation of aerodynamic forces on an aircraft |
| US8725470B1 (en) * | 2010-05-17 | 2014-05-13 | The United States of America as represented by the Administrator of the National Aeronautics & Space Administration (NASA) | Co-optimization of blunt body shapes for moving vehicles |
| CN104392075A (en) * | 2014-12-15 | 2015-03-04 | 中国飞机强度研究所 | Airfoil profile parametric modeling method |
-
2015
- 2015-04-22 CN CN201510198979.2A patent/CN104834772B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090157364A1 (en) * | 2007-12-18 | 2009-06-18 | Airbus Espana, S.L.. | Method and system for a quick calculation of aerodynamic forces on an aircraft |
| CN101458735A (en) * | 2008-12-31 | 2009-06-17 | 重庆大学 | Aerofoil with high lift-drag ratio |
| US8725470B1 (en) * | 2010-05-17 | 2014-05-13 | The United States of America as represented by the Administrator of the National Aeronautics & Space Administration (NASA) | Co-optimization of blunt body shapes for moving vehicles |
| CN104392075A (en) * | 2014-12-15 | 2015-03-04 | 中国飞机强度研究所 | Airfoil profile parametric modeling method |
Non-Patent Citations (2)
| Title |
|---|
| 陈杰: "基于神经网络方法的民用飞机先进气动力机翼设计研究", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
| 陈杰等: "基于SOM神经网络的超临界翼型设计", 《力学季刊》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107014451A (en) * | 2017-05-03 | 2017-08-04 | 东南大学 | The method of ultrasonic flow sensor coefficient is speculated based on generalized regression nerve networks |
| CN108733914A (en) * | 2018-05-17 | 2018-11-02 | 复旦大学 | Transonic airfoil Natural Laminar Flow delay based on artificial neural network turns to twist design method |
| DE102018210894A1 (en) * | 2018-07-03 | 2020-01-09 | Siemens Aktiengesellschaft | Design and manufacture of a turbomachine blade |
| WO2020112023A1 (en) * | 2018-11-26 | 2020-06-04 | Agency For Science, Technology And Research | Method and system for predicting performance in electronic design based on machine learning |
| CN110084511A (en) * | 2019-04-25 | 2019-08-02 | 广东工业大学 | A kind of unmanned plane matching method, device, equipment and readable storage medium storing program for executing |
| CN110084511B (en) * | 2019-04-25 | 2021-07-13 | 广东工业大学 | Unmanned aerial vehicle configuration method, device, equipment and readable storage medium |
| CN110321588A (en) * | 2019-05-10 | 2019-10-11 | 中车青岛四方车辆研究所有限公司 | Rail vehicle aerodynamic Drag Calculation method based on numerical simulation |
| CN112784508A (en) * | 2021-02-12 | 2021-05-11 | 西北工业大学 | Deep learning-based airfoil flow field rapid prediction method |
| CN113849910A (en) * | 2021-09-23 | 2021-12-28 | 四川大学 | Dropout-based BiLSTM network wing resistance coefficient prediction method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104834772B (en) | 2018-01-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104834772A (en) | Artificial-neural-network-based inverse design method for aircraft airfoils/wings | |
| CN109460566B (en) | Aerodynamic robust optimization design method for thick airfoil section on inner side of wind turbine blade | |
| CN104778327B (en) | Aircraft wing Optimization Design based on artificial neural network | |
| CN106055791B (en) | Aircraft overall situation Aerodynamic optimization method based on Predictor-Correcting Algorithm | |
| Leifsson et al. | Multi-fidelity design optimization of transonic airfoils using physics-based surrogate modeling and shape-preserving response prediction | |
| CN112016167B (en) | Aircraft aerodynamic shape design method and system based on simulation and optimization coupling | |
| CN106126860B (en) | A kind of hypersonic wing Robust Optimal Design for considering mismachining tolerance | |
| CN109711048A (en) | It is a kind of to consider pneumatic and structure polymorphic type design variable Wing design method | |
| CN103488847B (en) | A kind of aerodynamic shape optimization method based on Artificial neural network ensemble | |
| CN105183996A (en) | Surface element correction and grid beforehand self-adaption calculation method | |
| CN109460629A (en) | A kind of cooling fan performance optimization method based on approximate model method | |
| CN106845019A (en) | A kind of adaptive wing Airfoil Design method | |
| CN104573282A (en) | Aerodynamic optimum design method of airfoil profile of vertical axis wind turbine | |
| CN105718634A (en) | Airfoil robust optimization design method based on non-probability interval analysis model | |
| CN104978498A (en) | Adaptive method of biomolecule network topological structure | |
| CN104978449A (en) | Aerodynamic optimization method of leading edge slats position and trailing edge flap position of two-dimensional three-section airfoil profile | |
| CN105404743A (en) | Wind turbine airfoil design method combining B spline with curvature smooth continuity | |
| CN111814246A (en) | Wing section inverse design method based on generation countermeasure network | |
| CN116305523B (en) | Laminar flow airfoil optimization method based on discrete accompaniment | |
| CN115358001A (en) | Aerodynamic stealth comprehensive optimization method for front edge radius of flying wing layout aircraft | |
| CN104765927A (en) | Plane lift-enhancing device high-velocity and high-velocity comprehensive optimum design method based on multiple subjects | |
| CN105138766B (en) | The method of adding some points of hypersonic Aerodynamic Heating reduced-order model based on fuzzy clustering | |
| CN105117541B (en) | A kind of forward direction jig-shape optimum design method | |
| CN114021492B (en) | Supercritical airfoil buffeting optimization method based on neural network | |
| Barrett et al. | Airfoil shape design and optimization using multifidelity analysis and embedded inverse design |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180112 |