Optimization under worst case constraints—a new global multimodel search procedure
Tóm tắt
A new method is presented that combines heuristic global optimization and multi-model simulation for reliability based risk averse design. The so-called new stack ordering method is motivated from hydrogeology, where high-reliable groundwater management solutions are sought for with a demanding set of equally probable model alternatives. The idea is to only exploit a small subset of these model alternatives or realizations to approximate the objective function to reduce computational costs. The presented automatic procedure dynamically adjusts the subset online during the course of iterative optimization. The test with theoretical reliability based benchmark problems shows that the new method is efficient in regard to optimality and reliability of found solutions already with small subsets of all models. Compared with a previously presented first version of stack ordering, the presented generalized approach proves to be more robust, computationally efficient and of great potential for related problems in reliability based optimization and design. This conclusion is supported by the fact that the new variant requires about one fifth of the objective function evaluations of the older version in order to achieve the same level of reliability. We also show that these findings can be translated to real world problems by bench marking the performance on a well capture problem.
Tài liệu tham khảo
Arnst M, Ghanem R, Soize C (2010) Identification of bayesian posteriors for coefficients of chaos expansions. J Comput Phys 229(9):3134–3154. doi:10.1016/j.jcp.2009.12.033. http://www.sciencedirect.com/science/article/B6WHY-4Y34SWD-2/2/f6537e3705bb6566ef2e6717b0f8f7b5
Avigad G, Branke J (2008) Embedded evolutionary multi-objective optimization for worst case robustness. In: Ryan C, Keijzer M (eds) GECCO. ACM, pp 617–624
Avigad G, Coello CAC (2010) Highly reliable optimal solutions to multi-objective problems and their evolution by means of worst-case analysis. Eng Optim 42(12):1095–1117. http://www.informaworld.com/10.1080/03052151003668151
Azadivar F (1999) Simulation optimization methodologies. In: Proceedings of the 31st conference on winter simulation: simulation—a bridge to the future, vol 1 ACM, New York, WSC ’99, pp 93–100. http://doi.acm.org/10.1145/324138.324168
Bayer P, Bürger C, Finkel M (2004) Evolutionary algorithms for the optimization of advective control of contaminated zones. Water Resour Res 40:(1146):W06506. doi:10.1029/2003WR002675
Bayer P, Bürger C, Finkel M (2004) Computationally efficient stochastic optimization using multiple realizations. Adv Water Resour 31(2):399–417. doi:10.1016/j.advwatres.2007.09.004. http://www.sciencedirect.com/science/article/B6VCF-4PT0Y85-1/2/0b7d5cdbd9868f3a5127c7d306968808
Bayer P, de Paly M, Bürger CN (2010) Optimization of high-reliability-based hydrological design problems by robust automatic sampling of critical model realizations. Water Resour Res 46(5):W05504. doi:10.1029/2009WR008081
Binder K (1979) Monte Carlo methods in statistical physics / with contributions by K. Binder... [et al.]. In: Binder K (ed). Springer-Verlag, Berlin, New York
Cirpka OA, Bürger CM, Nowak W, Finkel M (2004) Uncertainty and data worth analysis for the hydraulic design of funnel-and-gate systems in heterogeneous aquifers. Water Resour Res 40(11):W11502. doi:10.1029/2004WR003352.
Debusschere BJ, Najm HN, Pébay PP, Knio OM, Ghanem RG, Maître OPL (2005) Numerical challenges in the use of polynomial chaos representations for stochastic processes. SIAM J Sci Comput 26:698–719. doi:10.1137/S1064827503427741
de Paly M, Schuetze N, Zell A (2010) Determining crop-production functions using multi-objective evolutionary algorithms. In: Proceedings of the IEEE congress on evolutionary computation (CEC), pp 1870–1877, Barcelona, Spain. doi: 10.1109/CEC.2010.5586147
Du X, Guo J, Beeram H (2008) Sequential optimization and reliability assessment for multidisciplinary systems design. Struct Multidiscip Optim 35:117–130. doi: 10.1007/s00158-007-0121-7
Fu MC, Glover F, April J (2005) Simulation optimization: a review, new developments, and applications. In: Winter simulation conference. ACM, pp 83–95.
Fink D (1997) A Compendium of Conjugate Priors. Tech. rep., Montana State Univeristy, URL http://www.johndcook.com/CompendiumOfConjugatePriors.pdf
Ghanem RG, Spanos PD (1991) Stochastic finite elements: a spectral approach. Springer-Verlag, New York
Hansen N, Ostermeier A (2001) Completely derandomized self-adaptation in evolution strategies. Evol Comput 9(2):159–195
Harbaugh A, Banta E, Hill M, McDonald M (2000) MODFLOW-2000, the U.S. Geological Survey modular ground water model, user guide to modularization concepts and the ground water flow process
Hasofer AM, Lind NC (1974) Exact and invariant second-moment code format. J Eng Mech Div 100(1):111–121
Hubbard WD (2007) How to measure anything finding the value of ‘intangibles’ in business. Wiley, Hoboken
Jeffreys H (1946) An invariant form for the prior probability in estimation problems. Proc Royal Soc Lond Ser A Math Phys Sci 186(1007):453–461. doi: 10.2307/97883
Kourakos G, Mantoglou A (2008) Remediation of heterogeneous aquifers based on multiobjective optimization and adaptive determination of critical realizations. Water Resour Res 44(12):W12408. doi:10.1029/2008WR007108
Krink T, Filipic B, Fogel G, Thomsen R (2004) Noisy optimization problems - a particular challenge for differential evolution? In: Proceedings of 2004 congress on evolutionary computation, pp 332–339. IEEE Press
Kronfeld M, Planatscher H, Zell A (2010) The EvA2 optimization framework. In: Blum C, Battiti R (eds) Learning and intelligent optimization conference, special session on software for optimization (LION-SWOP). Springer Verlag, Venice, Italy. no. 6073 in lecture notes in computer science, LNCS, pp 247–250. http://www.ra.cs.uni-tuebingen.de/publikationen/2010/Kron10Ev
Lin G, Tartakovsky A (2009) An efficient, high-order probabilistic collocation method on sparse grids for three-dimensional flow and solute transport in randomly heterogeneous porous media. Adv Water Resour 32(5):712–722. doi: 10.1016/j.advwatres.2008.09.003. http://www.sciencedirect.com/science/article/B6VCF-4TJ6FP4-1/2/7e4ecf6e3963323bede48fc4e9fa2fc3, dispersion in Porous Media
Lu Z, Zhang D (2005) A comparative study on uncertainty quantification for flow in randomly heterogeneous media using monte carlo simulations and conventional and kl-based moment-equation approaches. SIAM J Sci Comput 26:558–577. doi: 10.1137/S1064827503426826
Lu Z, Zhang D (2007) Stochastic simulations for flow in nonstationary randomly heterogeneous porous media using a kl-based moment-equation approach. Multiscale Model Simul 6(1):228–245. doi:10.1137/060665282. http://link.aip.org/link/?MMS/6/228/1
Mantoglou A, Kourakos G (2007) Optimal groundwater remediation under uncertainty using multi-objective optimization. Water Resour Manag 21:835–847. doi: 10.1007/s11269-006-9109-0
Morgan DR, Eheart JW, Valocchi AJ (1993) Aquifer remediation design under uncertainty using a new chance constrained programming technique. Water Resour Res 29(3):551–561. doi: 10.1029/92WR02130
Nicklow J, Reed P, Savic D, Dessalegne T, Harrell L, Chan-Hilton A, Karamouz M, Minsker B, Ostfeld A, Singh A, on Evolutionary Computation in Environmental EZATC, Engineering WR (2010) State of the art for genetic algorithms and beyond in water resources planning and management. J Water Resour Plan Manag 136(4):412–432. doi:10.1061/(ASCE)WR.1943-5452.0000053. http://link.aip.org/link/?QWR/136/412/1
Pollock DW (1994) Users guide for modpath/mod-path-plot, version a particle tracking post-processing package for modflow, the U.S. geological survey finite-difference ground-water flow
Price KV, Storn RM, Lampinen JA (2005) Differential evolution a practical approach to global optimization. Natural computing series. Springer-Verlag, Berlin
Singh A, Minsker BS (2008) Uncertainty-based multiobjective optimization of groundwater remediation design. Water Resour Res 44(2):W02404. doi:10.1029/2005WR004436
Smalley JB, Minsker BS, Goldberg DE (2000) Risk-based in situ bioremediation design using a noisy genetic algorithm. Water Resour Res 36(10):3043–3052. doi:10.1029/2000WR900191
Tvedt L (1990) Distribution of quadratic forms in normal space - application to structural reliability. J Eng Mech 116(6):1183–1197. doi:10.1061/(ASCE)0733-9399(1990)116:6(1183)
Valdebenito MA, Schuëller GI (2010) A survey on approaches for reliability-based optimization. Struct Multidiscip Optim 42:645–663. doi:10.1007/s00158-010-0518-6
Vu V (2010) Minimum weight design for toroidal pressure vessels using differential evolution and particle swarm optimization. Struct Multidiscip Optim 42:351–369. doi:10.1007/s00158-010-0494-x
Wagner BJ, Gorelick SM (1989) Reliable aquifer remediation in the presence of spatially variable hydraulic conductivity: from data to design. Water Resour Res 25(10):2211–2225. doi:10.1029/WR025i010p02211
Wu J, Zheng C, Chien CC, Zheng L (2006) A comparative study of Monte Carlo simple genetic algorithm and noisy genetic algorithm for cost-effective sampling network design under uncertainty. Adv Water Resour 29:899–911. doi:10.1016/j.advwatres.2005.08.005
Xi Z, Hu C, Youn B (2011) A comparative study of probability estimation methods for reliability analysis. Struct Multidiscip Optim 1–20. doi:10.1007/s00158-011-0656-5
Zhang D, Lu Z (2004) An efficient, high-order perturbation approach for flow in random porous media via karhunen-lo & 232;ve and polynomial expansions. J Comput Phys 194:773–794. doi:10.1016/j.jcp.2003.09.015. http://portal.acm.org/citation.cfm?id=1008428.1008445