An algorithm to determine sample sizes for optimization with artificial neural networks

This article presents an algorithm developed to determine the appropriate sample size for constructing accurate artificial neural networks as surrogate models in optimization problems. In the algorithm, two model evaluation methods—cross-validation and/or bootstrapping—are used to estimate the performance of various networks constructed with different sample sizes. The optimization of a CO2 capture process with aqueous amines is used as the case study to illustrate the application of the algorithm. The output of the algorithm—the network constructed using the appropriate sample size—is used in a process synthesis optimization problem to test its accuracy. The results show that the model evaluation methods are successful in identifying the general trends of the underlying model and that objective function value of the optimum solution calculated using the surrogate model is within 1% of the actual value. © 2012 American Institute of Chemical Engineers AIChE J, 59: 805–812, 2013

[1]  Daniel Ibraim Pires Atala,et al.  Optimisation of a continuous flash fermentation for butanol production using the response surface methodology , 2010 .

[2]  Gintaras V. Reklaitis,et al.  Simulation-based optimization with surrogate models - Application to supply chain management , 2005, Comput. Chem. Eng..

[3]  Christos T. Maravelias,et al.  Surrogate-Based Process Synthesis , 2010 .

[4]  V. Dua.  A mixed-integer programming approach for optimal configuration of artificial neural networks , 2010 .

[5]  Roberto Guardani,et al.  Neural network based approach for optimization of industrial chemical processes , 2000 .

[6]  Farrokh Mistree,et al.  Kriging Models for Global Approximation in Simulation-Based Multidisciplinary Design Optimization , 2001 .

[7]  Vivek Dua,et al.  Disaggregation-aggregation based model reduction for refinery-wide optimization , 2011, Comput. Chem. Eng..

[8]  M. Ierapetritou,et al.  A kriging method for the solution of nonlinear programs with black‐box functions , 2007 .

[9]  Jon C. Helton,et al.  Extension of Latin hypercube samples with correlated variables , 2008, Reliab. Eng. Syst. Saf..

[10]  Carolyn Conner Seepersad,et al.  Building Surrogate Models Based on Detailed and Approximate , 2004, DAC 2004.

[11]  Christos T. Maravelias,et al.  Surrogate‐based superstructure optimization framework , 2011 .

[12]  A. J. Yuste,et al.  A Neural Network Approach to Simulate Biodiesel Production from Waste Olive Oil , 2006 .

[13]  Honghai Wang,et al.  RSM optimization of the operating parameters for a butanol distillation column , 2012 .

[14]  Joos Vandewalle,et al.  Special issue on fundamental and information processing aspects of neurocomputing , 2002, Neurocomputing.

[15]  I A Basheer,et al.  Artificial neural networks: fundamentals, computing, design, and application. , 2000, Journal of microbiological methods.

[16]  R. M. Filho,et al.  Simulation and response surface analysis for the optimization of a three-phase catalytic slurry reactor , 2004 .

[17]  I. Grossmann,et al.  An algorithm for the use of surrogate models in modular flowsheet optimization , 2008 .

[18]  Raphael T. Haftka,et al.  Surrogate-based Analysis and Optimization , 2005 .

[19]  Selen Cremaschi,et al.  Optimization of CO2 capture process with aqueous amines using response surface methodology , 2011, Comput. Chem. Eng..

[20]  Arthur W. Westerberg,et al.  A review of process synthesis , 1981 .

[21]  I. Grossmann,et al.  A systematic modeling framework of superstructure optimization in process synthesis , 1999 .

[22]  Fabiano A.N. Fernandes,et al.  Optimization of Fischer‐Tropsch Synthesis Using Neural Networks , 2006 .

[23]  Barbara S. Minsker,et al.  Applying Dynamic Surrogate Models in Noisy Genetic Algorithms to Optimize Groundwater Remediation Designs , 2011 .

[24]  Andy J. Keane,et al.  Recent advances in surrogate-based optimization , 2009 .

[25]  Olivier Boutin,et al.  Process optimisation using the combination of simulation and experimental design approach: Application to wet air oxidation , 2011 .

[26]  George-Christopher Vosniakos,et al.  Optimizing feedforward artificial neural network architecture , 2007, Eng. Appl. Artif. Intell..