Hostname: page-component-78c5997874-s2hrs Total loading time: 0 Render date: 2024-11-19T13:31:58.917Z Has data issue: false hasContentIssue false
  • English
  • Français

Repetitively enhanced neural networks method for complex engineering design optimisation problems

Published online by Cambridge University Press:  27 January 2016

N. Van Nguyen ,
J-W. Lee ,
M. Tyan  and
S. Kim
N. Van Nguyen*
Affiliation:
Aerospace Information Engineering, Konkuk University, Seoul, South Korea
J-W. Lee*
Affiliation:
Aerospace Information Engineering, Konkuk University, Seoul, South Korea
M. Tyan
Affiliation:
Aerospace Information Engineering, Konkuk University, Seoul, South Korea
S. Kim
Affiliation:
Aerospace Information Engineering, Konkuk University, Seoul, South Korea
*
[email protected]
[email protected]
Get access Rights & Permissions [Opens in a new window]

Abstract

A Repetitively Enhanced Neural Networks (RENN) method is developed and presented for complex and implicit engineering design problems. The enhanced neural networks module constructs an accurate surrogate model and avoids over-fitting during neural networks training from supervised learning data. The optimiser is executed by the enhanced neural networks models to seek a tentative optimum point. It is repetitively added into the supervised learning data set to refine the surfaces until the RENN tolerance is reached. The RENN method demonstrates the effectiveness and feasibility for a 2D highly non-linear numerical example and the structure design of a two-member frame reaching a convergent solution at 10 and 15 iterations at the maximum error of 1% when compared with the exact solution. Then, the RENN method is applied for a long endurance unmanned aerial vehicle (UAV) aerofoil design optimisation. A Class/Shape function transformation (CST) geometry parameterisation method represents an accurate UAV aerofoil with ten geometry design variables. The high-fidelity analysis solver with structured mesh is used for a UAV aerofoil design problem. Using the RENN method, an optimal UAV aerofoil is obtained using 88 high fidelity evaluations at an error of 1·24%. The process reduces the computational time by 81·2% compared with the full high fidelity model. The optimal aerofoil shows a drag reduction of 10·8% in the cruise condition and an improvement in the maximum lift coefficient and stall angle-of-attack when compared with the baseline AG24 aerofoil.

Type
Research Article
Information
The Aeronautical Journal , Volume 119 , Issue 1220 , October 2015 , pp. 1253 - 1270
Copyright
Copyright © Royal Aeronautical Society 2015

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1.Changyu, S., Lixia, W. and Quan, L.Optimization of injection molding process parameters using combination of artifcial neural network and genetic algorithm, Sci Direct J, 2007, 183, (2-3), pp 412418.Google Scholar
2.Ozcelik, B. and Erzurumlu, T.Comparison of the warpage optimization in the plastic injection molding using ANOVA, neural network model and genetic algorithm, J Materials Processing Technology, 2006, 171, pp 437445.Google Scholar
3.Mok, S.L., Kwong, C.K. and Lau, W.SA hybrid neural network and genetic algorithm approach to the determination of initial process parameters for injection molding, Int J Advanced Manufacturing Tech, September 2001, 18, pp 404–409.Google Scholar
4.Park, K.H., Jun, S.O. and Baek, S.Met al, Reduced-order model with an artifcial neural network for aerostructural design optimization, J Aircr, July-August 2013, 50, (4).Google Scholar
5.Mazhar, F., Khan, A.M., Chaudhry, I.A. and Ahsan, M.On using neural networks in U AV structural design for CFD data fitting and classifcation, Aerospace Science and Technology, 2013, 30, pp 210225.Google Scholar
6.Athar Kharal, A. and Saleem, A.Neural networks based aerofoil generation for a given Cp using BezierPARSEC parameterization, Aerospace Science and Technology, 2012, 23, pp 330344.Google Scholar
7.Samy, I., Postlethwaite, I. and Gu, D.-W.Neural-Network-Based Fush Air Data Sensing System Demonstrated On A Mini Air Vehicle, J Aircr, January-February 2010, 47, (1).Google Scholar
8.Rai, M.M. and Madavan, N.K.Aerodynamic design using neural networks, AIAA J, 2000, 38, (1), pp 173–182.CrossRefGoogle Scholar
9.Saijal, K.K., Ganguli, R. and Viswamurthy, S.R.Optimization of helicopter rotor using polynomial and neural network metamodels, J Aircr, March-April 2011, 48, (2).Google Scholar
10.Patnaik, S.N., Coroneos, R.M., Guptill, J.D. and Hopkins, D.ASubsonic aircraft design optimization with neural network and regression approximators, J Aircr, September-October 2005, 42, (5).Google Scholar
11.Shyy, W., Papila, N., Vaidyanat han, R. and Tucker, K.Global design optimization for aerodynamics and rocket propulsion components, Progress in Aerospace Sciences, 2001, 37, pp 59–118.Google Scholar
12.Oroumieh, M.A.A., Malaek, S.M.B., Ashrafizaa deh, M. and Taheri, S.MAircraft design cycle time reduction using artifcial intelligence, Aerospace Science and Technology, 2013, 26, pp 244258.Google Scholar
13.Sinha, M., Ayilliat, R., Ajoy, K., Ghosh, K. and Misra, A.High angle-of-attack parameter estimation of cascaded fins using neural network, J Aircr, January-February 2013, 50.Google Scholar
14.Horn, J.F., Schmidt, E.M., Brian, R., Geiger, B.R. and Deangelo, M.P.Neural network-based trajectory optimization for unmanned aerial vehicles, J Guidance, Control, and Dynamics, March-April 2012, 35, (2).Google Scholar
15.Hacioglu, A.Fast evolutionary algorithm for aerofoil design via neural network, AIAA J, September 2007, 45, (9).Google Scholar
16.Queipo, N.V., Haftka, R.T., Shyy, W., Goel, T., Vaidynat han, R. and Tucker, P.K.Surrogate based analysis and optimization, Progress in Aerospace Sciences, January 2005, 41, (1), pp 1–28.Google Scholar
17.Forrester, A.I.J. and Keane, A.JRecent advances in surrogate-based optimization, Progress in Aerospace Sciences, January-April 2009, 45, 1-3, pp 50–79.Google Scholar
18.Koziel, S., Echeverra-Ciaurri, D. and Leifsson, L.Surrogate-based methods, in koziel, S. and Yang, X.S. (Eds) Computational Optimization, Methods and Algorithms, Series: Studies in Computational Intelligence, Springer-Verlag, pp 33–60, 2011.Google Scholar
19.Koziel, S. and Leifsson, L.Multi-Level Surrogate-Based Aerofoil Shape Optimization, 51st AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition, Grapevine, Texas, USA, 7-10 January, 2013.Google Scholar
20.Koziel, S. and Leifsson, L.Surrogate-based aerodynamic shape optimization by variable-resolution models, AIAA J, 2013, 51, (1), pp 94–106.Google Scholar
21.Koziel, S. and Leifsson, L.Knowledge-Based Aerofoil Shape Optimization Using Space Mapping, 30th AIAA Applied Aerodynamics Conference, New Orleans, Louisiana, USA, 25-28 June 2012.Google Scholar
22.S. I. Inc., Jmp(R) User Guide. 2nd Edition, Cary, NC: Sas Institute Inc, 2009.Google Scholar
23.Myers, R.H. and Montgomery, D.C.Response Surface Methodology: Process and Product Optimization Using Designed Experiments, Wiley-Interscience, 1995.Google Scholar
24.Hagan, M.T., Demuth, H.B. and Beale, M.H.Neural Network Design, PWS, 1996.Google Scholar
25.Nguyen, D. and Widrow, B.Improving the learning speed of 2-layer neural networks by choosing initial values of the adaptive weights, Int Joint Conf Neural Networks, June 1990, 3, pp 2126.Google Scholar
26.MacKay, D.J.C.A practical Bayesian process for backprop networks, Neural Computation J, 1992, 4, pp 448472.Google Scholar
27. Matlab 2013, Neural Networks Toolbox, http://www.mathworks.co.kr/Google Scholar
28.Goldberg, D.E.Genetic Algorithms in Search, Optimization, and Machine Learning, Addison-Webley-Professional, 1989.Google Scholar
29.Arora, J.SIntroduction to Optimum Design, 2nd ed, Elsevier Academic Press, 2004.Google Scholar
30.Jeon, K.-S., Lee, J-W. and Byun, Y.-H.Repetitive Response Surface Enhancement Techniques through Design Space Transformation and Sub-Optimization ICAS 25th Int Congress of the Aeronautical Sci, Hamburg, Germany, September 2006.Google Scholar
31.Kulfan, B.M.A universal parametric geometry representation method CST, J Aircr, JanuaryFebruary 2008, 45, (1).Google Scholar
32.Vu, N.A., Lee, J-W. and Shu, J.Aerodynamic design optimization of helicopter rotor blades including aerofoil shape for hover performance, Chinese J Aeronautics, 2013, 26, (1), pp 1–8.Google Scholar
33.Azamatov, A., Lee, J-W. and Byun, Y.-H.Comprehensive aircraft confguration design tool for integrated product and process development, Advances in Engineering Software, 2011, 42, pp 35–49.Google Scholar
34. ANSYS FLUENT flow modelling and simulation software, Ansys Inc.Google Scholar
35.Drela, M. Xfoil program, available: http://web.mit.edu/drela/Public/web/xfoil/Google Scholar
36. Javafoil Analysis of Aerofoils, http://www.mh-aerotools.de/aerofoils/javafoil.htmGoogle Scholar
37.Williamson, A.G., McGranahan, B.D., Broughton, B.A., Deters, R.W., Brandt, J.B. and Selig, M.SSummary of low-speed aerofoil data, Department of Aerospace Engineering, University of Illinois at Urbana-Champaign, Illinois, USA, 2012.Google Scholar
38. Gridgen Pointwise, http://www.pointwise.com/gridgen/Google Scholar
39.Tyan, M., Van Nguyen, N. and Lee, J.-W.Improving variable fdelity modelling by exploring global design space and radial basis functions networks for aerofoil design, Engineering Optimization Journal, Taylor & Francis, DOI: 10.1080/0305215X.2014.941290, 2014.Google Scholar