ABSTRACT
This paper presents a data driven, surrogate based optimization algorithm that uses a sequential approximate optimization (SAO) framework and a statistical sampling approach. The biggest concern in using the SAO framework based on statistical sampling is the generation of the required database. A data driven approach is proposed to tackle this situation, where the trick is to run the expensive simulation if and only if a nearby data point does not exist in the cumulatively growing database. Results show that the proposed methodology dramatically reduces the total number of calls to the expensive simulation runs during the optimization process. Utilizing parallel or distributed computing for generating the required database can further reduce the optimization time, as can a high performance parallel computing back-end for each single simulation run.
- Bramley, R.; Gannon, D.; Stuckey, T.; Villacis, J.; Akman, E.; Balasubramanian, J.; Breg, F.; Diwan, S.; Govindaraju, M. (1998), "The linear system analyzer, technical Report TR-511", Dept. of Computer Sci., Indiana University, Bloomington, IN.Google Scholar
- Boisvert, R. F.; Rice, J. R. (1985). Solving Elliptic Problems Using ELLPACK, Springer-Verlag: New York, NY. Google ScholarDigital Library
- Gallopoulos, E.; Houstis, E.; Rice, J. R. (1994). "Computer as thinker/doer: Problem solving environments for computational science", IEEE Computational Sci and Eng, 1, 11--23. Google ScholarDigital Library
- Goel, A.; Phanouriou, C.; Kamke, F. A.; Ribbens, C. J.; Shaffer, C. A.; Watson, L. T. (1999). "WBCSim: a prototype problem solving environment for wood-based composites simulations", Engineering with Computers, 15, 198--210.Google ScholarCross Ref
- Giunta, A. A.; Richards, M. D.; Cyr, E. C.; Swiler, L. P.; Brown, S. L.; Eldred, M. S. (2006). Surfpack Version 1.0 User's Manual, Sandia National Laboratories, Albuquerque, NM, USA.Google Scholar
- Giunta, A. A.; Swiler, L. P.; Brown, S. L.; Eldred, M. S.; Richards, M. D.; Cyr, E. C. (2006). "The Surfpack software library for surrogate modeling of sparse irregularly spaced multidimensional data." In 11th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, Portsmouth, VA. AIAA-2006-7049.Google Scholar
- Houstis, E.; Gallopoulos, E.; Bramley, R.; Rice, J. R. (1997). "Problem solving environments for computational science", IEEE Computational Sci and Eng, 4, 18--21. Google ScholarDigital Library
- Kamke, F. A.; Wilson, J. B. (1985). "Computer simulation of a rotary dryer: retention time", American Institute of Chemical Engineers Journal, 32, 263--268.Google ScholarCross Ref
- Kamke, F. A.; Wilson, J. B. (1985). "Computer simulation of a rotary dryer: heat and mass transfer", American Institute of Chemical Engineers Journal, 32, 269--275.Google ScholarCross Ref
- Pérez, V. M.; Renaud, J. E.; Watson, L. T. (2001). "Adaptive Experimental Design for Construction of Response Surface Approximation", AIAA Journal, 40, 2495--2503.Google ScholarCross Ref
- Pérez, V. M.; Renaud, J. E.; Watson, L. T. (2008). "Reduced sampling for construction of quadratic response surface approximations using adaptive experimental design", Engineering Computations, 25, 764--782.Google ScholarCross Ref
- Pérez, V. M.; Renaud, J. E.; Gano, S. E. (2000). "Constructing variable fidelity response surface approximations in the usable feasible region." In Proceedings of the 8th AIAA/NASA/USAF Multidisciplinary Analysis & Optimization Symposium, Long Beach, CA. AIAA 2000--4888.Google Scholar
- Pérez, V. M.; Renaud, J. E. (2000). "Decoupling the design sampling region from the trust region in approximate optimization", Proceedings of the International Mechanical Engineering Congress and Exposition, American Society of Mechanical Engineers, 63, 205--214.Google Scholar
- Rodríguez, J. F.; Perez, V. M.; Padmanabhan, D.; Renaud, J. E. (2001). "Sequential approximate optimization using variable fidelity response surface approximations", Structural and Multidisciplinary Optimization, 22, 24--34.Google ScholarDigital Library
- Rodríguez, J. F.; Renaud, J. E.; Watson, L. T. (1998). "Trust region augmented Lagrangian methods for sequential response surface approximation and optimization", Journal of Mechanical Design, 120, 58--66.Google ScholarCross Ref
- Rodríguez, J. F.; Renaud, J. E.; Watson, L. T. (1998). "Convergence of trust region augmented Lagrangian methods using variable fidelity approximation data", Structural optimization, 15, 141--156.Google Scholar
- Shu, J.; Watson, L. T.; Ramakrishnan, N.; Kamke, F. A.; Zombori, B. G. (2004). "An experiment management component for the WBCSim problem solving environment", Advances in Engineering Software, 35, 115--123. Google ScholarDigital Library
- Shu, J.; Watson, L. T.; Zombori, B. G.; Kamke, F. A. (2006). "WBCSim: an environment for modeling wood-based composites manufacture", Engineering with Computers, 21, 259--271. Google ScholarDigital Library
- Thacker, W. I.; Zhang, J.; Watson, L. T.; Birch, J. B.; Iyer, M. A.; Barry, M. W. (2009), "Algorithm XXX: SHEPPACK: Modified Shepard algorithm for interpolation of scattered multivariate data", Technical Report TR 09--13, Department of Computer Science, Virginia Polytechnic Institute & State University, Blacksburg, VA 24061.Google Scholar
Recommendations
Data driven surrogate-based optimization in the problem solving environment WBCSim
Large scale, multidisciplinary, engineering designs are always difficult due to the complexity and dimensionality of these problems. Direct coupling between the analysis codes and the optimization routines can be prohibitively time consuming due to the ...
A New Sequential Approximate Optimization Approach Using Radial Basis Functions for Engineering Optimization
ICIRA 2015: Proceedings of the 8th International Conference on Intelligent Robotics and Applications - Volume 9246For most engineering optimization problems, it is difficult to find the global optimum due to the unaffordable computational cost. To overcome this difficulty, a new sequential approximate optimization approach using radial basis functions is proposed ...
Solving the multi-stage portfolio optimization problem with a novel particle swarm optimization
Research highlights We propose a drift particle swarm optimization (DPSO) algorithm. DPSO is applied to solve multi-stage portfolio optimization (MSPO) problems. MSPO problems with data from S&P 100 index are tested by DPSO and other algorithms. DPSO ...
Comments