Surrogate assisted optimization for parameters tuning of a separator in a tapioca starch process

Main Article Content

Sompoch Charoensuk
Thana Radpukdee
Nantiwat Pholdee

บทคัดย่อ

This study employs a single-objective differential evolution optimizer for surrogate-assisted optimization. A separator plant is modeled using five surrogate approaches for process optimization. The optimization objective is defined as minimizing starch loss relative to the Baume output of starch milk, while adhering to the lower and upper bounds of design parameters, including pulp input, Baume input, and regulated valve flow control. These bounds also support the plant controller's original design parameters. The optimal parameters were implemented and their real and numerical performances compared to validate the proposed method. The results demonstrate that the suitable surrogate model for the separator plant can be identified and that the optimal characteristics can be achieved, providing practical benefits to the process.

Article Details

How to Cite
1.
Charoensuk S, Radpukdee T, Pholdee N. Surrogate assisted optimization for parameters tuning of a separator in a tapioca starch process. featkku [อินเทอร์เน็ต]. 26 ธันวาคม 2024 [อ้างถึง 23 มกราคม 2025];10(2):51-67. available at: https://ph02.tci-thaijo.org/index.php/featkku/article/view/255914
บท
บทความวิจัย

References

Alexander I, Forrester J, András S, Andy J. Keane. Engineering Design via Surrogate Modelling: A Practical Guide, University of Southampton: UK: 2008. John Wiley & Sons Ltd.

Ali W, Khan M. S, Qyyum, M. A, Lee M. Surrogate-assisted modeling and optimization of a natural-gas liquefaction plant. Computers and Chemical Engineering 2018; 118 : 132-142.

Bagheri S, Reinicke U, Anders D, Konen W. Surrogate-assisted optimization for augmentation of finite element techniques. Journal of Computational Science 2021; 54: 101427.

Chen X, Wu K, Bai A, Masuku C. M, Niederberger J, Liporace, F. S, et al. Real-time refinery optimization with reduced-order fluidized catalytic cracker model and surrogate-based trust region filter method. Computers and Chemical Engineering 2021; 153: 107455.

Franzoi R. E, Kelly J. D, Menezes B. C, Swartz C. L. E. An adaptive sampling surrogate model building framework for the optimization of reaction systems. Computers and Chemical Engineering 2021; 152: 107371.

Hu A, Liao H, Guan W, Dong J, Qian X. Support vector machine model based on OTSU segmentation algorithm in diagnosing bronchiectasis with chronic airway infections. J Radiat Res Appl Sci 2023; 16: 100500. https://doi.org/10.1016/j.jrras.2022.100500.

Rojas-López A. G, Villarreal-Cervantes M. G, Rodríguez-Molina A. Surrogate indirect adaptive controller tuning based on polynomial response surface method and bioinspired optimization: Application to the brushless direct current motor controller. Expert Systems with Applications 2024; 245: 123070.

Stavros P. Adam, Stamatios-Aggelos N. Alexandropoulos, Panos M. Pardalos, and Michael N. Vrahatis (2019). No Free Lunch Theorem: A Review. Research Gate DOI: 10.1007/978-3-030-12767-1_5.

Sujin Bureerat. Mechanical System Engineering Optimization 1-2. Khon Kaen Univercity; 2013.

Wencheng J, Trevor A. A, Christine D, Ghanashyam N, Nicolas S, Travis L. M, et al. Machine-learning-assisted high-temperature reservoir thermal energy storage optimization, Renewable Energy 197 (2022) :384–397.

Ong Y. S, Zhou Z, Lim D. Curse and blessing of uncertainty in evolutionary algorithm using approximation. In 2006 IEEE international conference on evolutionary computation; 2006 July. IEEE; pp. 2928-2935.