Dynamic modelling and thermal control of an integrated membrane distillation/heat pump process for hypersaline brines treatment

Simple item page
dc.catalogadoryvc
dc.contributor.advisorPérez Correa, José Ricardo
dc.contributor.advisorDíaz Quezada, Simón Diego
dc.contributor.authorValenzuela Álvarez, Benjamín Enrique
dc.contributor.otherPontificia Universidad Católica de Chile. Escuela de Ingeniería
dc.date.accessioned2025-10-28T19:39:25Z
dc.date.available2025-10-28T19:39:25Z
dc.date.issued2025
dc.descriptionTesis (Magíster en Ciencias de la Ingeniería)--Pontificia Universidad Católica de Chile, 2025
dc.description.abstractControl over membrane distillation-crystallization (MDCr) systems has gained considerable interest due to the benefits this technology offers in terms of water recovery and the concentration of hypersaline brines. This work presents the dynamic modeling and thermal control of a membrane distillation system integrated with a heat pump for energy recovery, designed for the treatment of hypersaline briners. The proposed fist-principles model, developed in MATLAB/Simulink represents an early-stage digital twin of a pilot-scale membrane distillation plant, incorporating mass and heat transfer phenomena across the membrane module and associated equipment. This system is currently under construction at the Advanced Mining Technology Center (AMTC), supporting the model’s relevance to real-world implementation and design workflows. Two control strategies were assessed to regulate the membrane and permeate temperatures: (i) a PID controller tuned using classical methods (Ziegler–Nichols and IMC-PI), and (ii) a Reinforcement Learning (RL) -based controller using a Deep Deterministic Policy Gradient (DDPG) agent. Both strategies were tested under realistic disturbances and structural variations in the membrane module to evaluate their robustness and adaptability. The PID control strategy resulted in energy savings of up to 45%,without significant losses in water recovery. The RL agent, trained for approximately two hours, achieved acceptable control performance, though it was outperformed by the PID strategy under the conditions tested. Results showed that the PID method outperformed the implemented RL agent by 88%. The results indicate that the PID control strategy outperformed in its ability to regulate the temperature of both tanks, while the RL controller showed potential to be used in this context. However, its performance was limited by the available computational. esources during training. This study contributes to advancing the technological maturity of membrane distillation systems by demonstrating model-based control design workflows aimed at improving scalability and energy efficiency.
dc.fechaingreso.objetodigital2025-10-28
dc.format.extent70 páginas
dc.fuente.origenSRIA
dc.identifier.urihttps://repositorio.uc.cl/handle/11534/106383
dc.information.autorucEscuela de Ingeniería; Valenzuela Álvarez, Benjamín Enrique; S/I; 1087325
dc.information.autorucEscuela de Ingeniería; Pérez Correa, José Ricardo; 0000-0002-1278-7782; 100130
dc.information.autorucEscuela de Ingeniería; Díaz Quezada, Simón Diego; S/I; 250656
dc.language.isoen
dc.nota.accesocontenido completo
dc.rightsacceso abierto
dc.subject.ddc620
dc.subject.deweyIngenieríaes_ES
dc.subject.ods07 Affordable and clean energy
dc.subject.odspa07 Energía asequible y no contaminante
dc.titleDynamic modelling and thermal control of an integrated membrane distillation/heat pump process for hypersaline brines treatment
dc.typetesis de maestría
sipa.codpersvinculados1087325
sipa.codpersvinculados100130
sipa.codpersvinculados250656
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
TESIS_BValenzuela.pdf
Size:
2.44 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
1.98 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
ING Tesis magíster