Advertisement

Learning to Set Up Numerical Optimizations of Engineering Designs

  • Mark Schwabacher
  • Thomas Ellman
  • Haym Hirsh
Part of the Massive Computing book series (MACO, volume 3)

Abstract

Gradient-based numerical optimization of complex engineering designs offers the promise of rapidly producing better designs. However, such methods generally assume that the objective function and constraint functions are continuous, smooth, and defined everywhere. Unfortunately, realistic simulators tend to violate these assumptions, making optimization unreliable. Several decisions that need to be made in setting up an optimization, such as the choice of a starting prototype, and the choice of a formulation of the search space, can make a difference in how reliable the optimization is. Machine learning can help by making these choices based on the results of previous optimizations. We demonstrate this idea by using machine learning for four parts of the optimization setup problem: selecting a starting prototype from a database of prototypes, synthesizing a new starting prototype, predicting which design goals are achievable, and selecting a formulation of the search space. We use standard tree-induction algorithms (C4.5 and CART). We present results in two realistic engineering domains: racing yachts, and supersonic aircraft. Our experimental results show that using inductive learning to make setup decisions improves both the speed and the reliability of design optimization.

Keywords

Search Space Design Goal Soft Constraint Inductive Learning Design Library 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Bhatta, S. and Goel, A., Model-based design indexing and index learning in engineering design, in Working Notes of the IJCAI Workshop on Machine Learning in Engineering, 1995.Google Scholar
  2. Bouchard, E. E., Kidwell, G. H., and Rogan, J. E., The Application of Artificial Intelligence Technology to Aeronautical System Design, in AIAA/AHS/ASEE Aircraft Design Systems and Operations Meeting,Atlanta, Georgia, 1988, AIAA-88–4426.Google Scholar
  3. Breiman, L., Classification And Regression Trees, Belmont, Calif.: Wadsworth International Group, 1984.zbMATHGoogle Scholar
  4. Cerbone, G., Machine learning in engineering: Techniques to speed up numerical optimization, Technical Report 92–30–09, Oregon State University Department of Computer Science, 1992, Ph.D. Thesis.Google Scholar
  5. Char, B., Geddes, K., Gonnet, G., Leong, B., Monagan, M., and Watt, S., First Leaves: A Tutorial Introduction to Maple V, Springer-Verlag and Waterloo Maple Publishing, 1992.zbMATHGoogle Scholar
  6. Choy, J. and Agogino, A., SYMON: Automated Symbolic Monotonicity Analysis System for Qualitative Design Optimization, in Proceedings ASME International Computers in Engineering Conference, 1986.Google Scholar
  7. Clark, P. and Matwin, S., Using qualitative models to guide inductive learning, in Proceedings of the tenth international machine learning conference, 49–56, Morgan Kaufmann, 1993.Google Scholar
  8. Ellman, T., Keane, J., and Schwabacher, M., The Rutgers CAP Project Design Associate, Technical Report CAP-TR-7, Department of Computer Science, Rutgers University, New Brunswick, NJ, 1992, http://ftp.cs.rutgers.edu/pub/technical-reports/cap-tr-7.ps.Z.Google Scholar
  9. Ellman, T., Keane, J., and Schwabacher, M., Intelligent Model Selection for Hillclimbing Search in Computer-Aided Design, in Proceedings of the Eleventh National Conference on Artificial Intelligence, 594–599, Washington, DC: MIT Press, Cambridge, MA, 1993.Google Scholar
  10. Ellman, T. and Schwabacher, M., Abstraction and Decomposition in Hillclimbing Design Optimization, Technical Report CAP-TR-14, Department of Computer Science, Rutgers University, New Brunswick, NJ, 1993, http://ftp.cs.rutgers.edu/pub/technical-reports/cap-tr-14.ps.Z.Google Scholar
  11. Gelsey, A., Schwabacher, M., and Smith, D., Using Modeling Knowledge to Guide Design Space Search, AI Journal, 101 (1–2), 35–62, 1998.zbMATHGoogle Scholar
  12. Gelsey, A., Smith, D., Schwabacher, M., Rasheed, K., and Miyake, K., A Search Space Toolkit: SST, Decision Support Systems, 18, 341–356, 1996.CrossRefGoogle Scholar
  13. Hoeltzel, D. and Chieng, W., Statistical Machine Learning for the Cognitive Selection of Nonlinear Programming Algorithms in Engineering Design Optimization, in Advances in Design Automation, Boston, MA, 1987.Google Scholar
  14. IYRU, The Rating Rule and Measurement Instructions of the International Twelve Metre Class, International Yacht Racing Union, 1985.Google Scholar
  15. Kolodner, J., Case-Based Reasoning, San Mateo, CA: Morgan Kaufmann Publishers, 1993.CrossRefzbMATHGoogle Scholar
  16. Lawrence, C., Zhou, J., and Tits, A., User’s Guide for CFSQP Version 2.3: A C Code for Solving (Large Scale) Constrained Nonlinear (Minimax) Optimization Problems, Generating Iterates Satisfying All Inequality Constraints, Technical Report TR-94–16r1, Institute for Systems Research, University of Maryland, College Park, MD, 1995.Google Scholar
  17. Letcher, J., The Aero/Hydro VPP Manual, Southwest Harbor, ME: Aero/Hydro, Inc., 1991.Google Scholar
  18. Letcher, J., Marshall, J., Oliver, J., and Salvesen, N., Stars and Stripes, Scientific American, 257 (2), 1987.Google Scholar
  19. Murthy, S., Kasif, S., Salzberg, S., and Beigel, R., A System for Induction of Oblique Decision Trees, Journal of Artificial Intelligence Research, 2, 1–32, 1994.zbMATHGoogle Scholar
  20. Orelup, M. E, Dixon, J. R., Cohen, P. R., and Simmons, M. K., Dominic II: Meta-Level Control in Iterative Redesign, in Proceedings of the National Conference on Artificial Intelligence, 2530, St. Paul, MN: MIT Press, Cambridge, MA, 1988.Google Scholar
  21. Papalambros, P. and Wilde, J., Principles of Optimal Design, New York, NY: Cambridge University Press, 1988.zbMATHGoogle Scholar
  22. Powell, D., Inter-GEN: A Hybrid Approach to Engineering Design Optimization, Ph.D. thesis, Rensselaer Polytechnic Institute Department of Computer Science, Troy, NY, 1990.Google Scholar
  23. Powell, D. and Skolnick, M., Using genetic algorithms in engineering design optimization with non-linear constraints, in Proceedings of the Fifth International Conference on Genetic Algorithms, 424–431, Univeristy of Illinois at Urbana-Champaign: Morgan Kaufmann, Los Altos, CA, 1993.Google Scholar
  24. Quinlan, J. R., Learning logical definitions from relations, Machine Learning, 5, 239–266, 1990.Google Scholar
  25. Quinlan, J. R., C4. 5: Programs for Machine Learning, San Mateo, CA: Morgan Kaufmann, 1993.Google Scholar
  26. Ramachandran, N., Langrana, N., Steinberg, L., and Jamalabad, V., Initial Design Strategies for Iterative Design, Research in Engineering Design, 4, 159–169, 1992.CrossRefGoogle Scholar
  27. Rasheed, K. and Hirsh, H., Learning to be Selective in Genetic-Algorithm-Based Design Opti- mization, Artificial Intelligence for Engineering Design, Analysis, and Manufacturing, 13, 1999.Google Scholar
  28. Rogers, D. and Adams, J., Mathematical elements for computer graphics, McGraw-Hill, second edition, 1990.Google Scholar
  29. Schwabacher, M., The Use of Artificial Intelligence to Improve the Numerical Optimization of Complex Engineering Designs, Technical Report HPCD-TR-45, Department of Computer Science, Rutgers University, New Brunswick, NJ, 1996, Ph.D. Thesis. http://www.cs.rutgers.edu/rschwabacithesis.html.
  30. Schwabacher, M. and Gelsey, A., Intelligent Gradient-Based Search of Incompletely Defined Design Spaces, Artificial Intelligence for Engineering Design, Analysis and Manufacturing, 11 (3), 199–210, 1997.Google Scholar
  31. Sycara, K. and Navinchandra, D., Retrieval Strategies in a Case-Based Design System, in C. Tong and D. Sriram, editors, Artificial Intelligence in Engineering Design (Volume II), 145 —164, New York, NY: Academic Press, 1992.Google Scholar
  32. Tong, S. S., Coupling Symbolic Manipulation and Numerical Simulation for Complex Engineering Designs, in International Association of Mathematics and Computers in Simulation Conference on Expert Systems for Numerical Computing, Purdue University, West Lafayette, IN, 1988.Google Scholar
  33. Tong, S. S., Powell, D., and Goel, S., Integration of Artificial Intelligence and Numerical Optimization Techniques for the Design of Complex Aerospace Systems, in 1992 Aerospace Design Conference,Irvine, CA, 1992, AIAA-92–1189.Google Scholar
  34. Williams, B. and Cagan, J., Activity Analysis: The Qualitative Analysis of Stationary Points for Optimal Reasoning, in Proceedings of the Twelfth National Conference on Artificial Intelligence, 1217–1223, Seattle, WA: MIT Press, 1994.Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2001

Authors and Affiliations

  1. 1.Computer Science DepartmentRutgers, The State University of New JerseyNew BrunswickUSA

Personalised recommendations