Hybrid analytical surrogate-based process optimization via Bayesian symbolic regression
Por favor, use este identificador para citar o enlazar este ítem:
http://hdl.handle.net/10045/139506
Título: | Hybrid analytical surrogate-based process optimization via Bayesian symbolic regression |
---|---|
Autor/es: | Jog, Sachin | Vázquez, Daniel | Santos, Lucas F. | Caballero, José A. | Guillén Gosálbez, Gonzalo |
Grupo/s de investigación o GITE: | Computer Optimization of Chemical Engineering Processes and Technologies (CONCEPT) |
Centro, Departamento o Servicio: | Universidad de Alicante. Departamento de Ingeniería Química | Universidad de Alicante. Instituto Universitario de Ingeniería de los Procesos Químicos |
Palabras clave: | Process optimization | Hybrid surrogate models | Black-box surrogate models | Bayesian symbolic regression |
Fecha de publicación: | 20-dic-2023 |
Editor: | Elsevier |
Cita bibliográfica: | Computers & Chemical Engineering. 2024, 182: 108563. https://doi.org/10.1016/j.compchemeng.2023.108563 |
Resumen: | Modular chemical process simulators are widespread in chemical industries to design and optimize production processes with sufficient accuracy. However, convergence issues and entrapment in local optima during process optimization are still challenges to overcome. To circumvent them, surrogate models of first principles simulations have attracted attention as they are easier to handle, with hybrid surrogates combining data-driven surrogate models with mechanistic equations becoming particularly appealing. In this context, this work explores the use of Bayesian symbolic regression to construct and globally optimize hybrid analytical surrogate models of process flowsheets, where some units are approximated with tailored analytical expressions rather than with neural networks or Gaussian processes, which are harder to globally optimize. Comparing with other prevalent black-box surrogate modeling & optimization approaches, such as kriging and Bayesian optimization, we find that our approach can find better solutions than using pure black-box methodologies, yet model building is much more computationally demanding. |
Patrocinador/es: | The authors would like to acknowledge the financial support from the Swiss National Science Foundation (Project LEARN-D, MINT 200021_214877). JAC also acknowledges the Spanish Ministerio de Ciencia y Innovación for the financial support under project PID2021-124139NB-C21. |
URI: | http://hdl.handle.net/10045/139506 |
ISSN: | 0098-1354 (Print) | 1873-4375 (Online) |
DOI: | 10.1016/j.compchemeng.2023.108563 |
Idioma: | eng |
Tipo: | info:eu-repo/semantics/article |
Derechos: | © 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). |
Revisión científica: | si |
Versión del editor: | https://doi.org/10.1016/j.compchemeng.2023.108563 |
Aparece en las colecciones: | INV - CONCEPT - Artículos de Revistas |
Archivos en este ítem:
Archivo | Descripción | Tamaño | Formato | |
---|---|---|---|---|
Jog_etal_2024_ComputChemEng.pdf | 1,47 MB | Adobe PDF | Abrir Vista previa | |
Todos los documentos en RUA están protegidos por derechos de autor. Algunos derechos reservados.