Commenced in January 2007
Frequency: Monthly
Edition: International
Paper Count: 30075
Improvement of Central Composite Design in Modeling and Optimization of Simulation Experiments

Authors: A. Nuchitprasittichai, N. Lerdritsirikoon, T. Khamsing

Abstract:

Simulation modeling can be used to solve real world problems. It provides an understanding of a complex system. To develop a simplified model of process simulation, a suitable experimental design is required to be able to capture surface characteristics. This paper presents the experimental design and algorithm used to model the process simulation for optimization problem. The CO2 liquefaction based on external refrigeration with two refrigeration circuits was used as a simulation case study. Latin Hypercube Sampling (LHS) was purposed to combine with existing Central Composite Design (CCD) samples to improve the performance of CCD in generating the second order model of the system. The second order model was then used as the objective function of the optimization problem. The results showed that adding LHS samples to CCD samples can help capture surface curvature characteristics. Suitable number of LHS sample points should be considered in order to get an accurate nonlinear model with minimum number of simulation experiments.

Keywords: Central composite design, CO2 liquefaction, Latin Hypercube Sampling, simulation – based optimization.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.3299525

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 110

References:


[1] A. C. Dimian, C. S. Bildea, and A. A. Kiss, “Chapter 2 – Introduction in Process Simulation,” Computer Aided Chemical Engineering, vol. 35, pp. 35 – 71, 2014.
[2] Y. Ma, J. Weng, Z. Shao, X. Chen, L. Zhu, Y. Zhao, and G. O. Young, “Parallel Computation Method for Solving Large Scale Equation – oriented Models,” Computer Aided Chemical Engineering, vol. 37, pp. 239 – 244, 2015
[3] A. M. Law “A tutorial on design of experiments for simulation modelling” Proceedings of the 2014 Winter Simulation Conference.
[4] B. G. M. Husslage, G. Rennen, E. R. v. Dam, and D. d. Hertog, “Space-filling Latin hypercube designs for computer experiments,” Optimization and Engineering, vol. 12, pp. 611–630, 2011.
[5] L. E. Øi, N. Eldrup, U. Adhikari, M. H. Bentsen, J. L. Badalge, S. Yang, “Simulation and cost comparison of CO2 liquefaction,” Energy Procedia, vol. 86, pp. 500 – 510, 2016.
[6] R. Turton, R. C. Bailie, W. B. Whiting, and J. A. Shaeiwitz, Analysis, synthesis and design of chemical processes, 2nd ed., Prentice Hall, 2003.
[7] A. Nuchitprasittichai, and S. Cremaschi, “Optimization of CO2 capture process with aqueous amines using response surface methodology,” vol.35, pp. 1521 – 1531, 2011.