Please use this identifier to cite or link to this item: https://scholarbank.nus.edu.sg/handle/10635/231564
Title: PERFORMANCE PREDICTION OF MULTILAYER ISOPOROUS MEMBRANE THROUGH PHYSICS-BASED NEURAL NETWORK FOR AIR FILTRATION APPLICATIONS
Authors: ONG SHI KE
ORCID iD:   orcid.org/0000-0002-2095-1351
Keywords: CFD, Isoporous membrane, Neural network, Monte Carlo simulation, sensitivity analysis, air filtration
Issue Date: 6-May-2022
Citation: ONG SHI KE (2022-05-06). PERFORMANCE PREDICTION OF MULTILAYER ISOPOROUS MEMBRANE THROUGH PHYSICS-BASED NEURAL NETWORK FOR AIR FILTRATION APPLICATIONS. ScholarBank@NUS Repository.
Abstract: Designing a filtration medium requires balancing high filtration efficiency and low energy requirement (i.e., low pressure drop). Unlike fibrous and random porous membranes, the pore architecture of isoporous membranes is deterministic, can be precisely controlled as per design, and is characterized by a periodic array of parallel pores with narrow pore size distribution. Although isoporous membrane allows for a deterministic approach towards targeted design and optimization of multiscale pore architecture, it is challenging to obtain free-standing and residual-layer-free isoporous membrane with submicron pores. As such, there exist an under-explored design space of multiscale pore architecture for the concurrent optimization of pressure drop and filtration efficiency. Thus, the goal of this thesis is to develop physics-based computational design models, validated with experiments, to aid in the deterministic approach of designing and optimizing the pore architecture of single and multilayer isoporous membrane to obtain low pressure drop and high filtration efficiency. Generally, artificial neural networks (ANNs) were trained with experimentally validated physics-based simulations where the incompressible, isothermal, laminar flow was solved. In this work, both single-layer and multilayer systems were investigated.
URI: https://scholarbank.nus.edu.sg/handle/10635/231564
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
Training data.xlsx297.51 kBMicrosoft Excel XML

OPEN

NoneView/Download
OngSK.pdf8.94 MBAdobe PDF

OPEN

NoneView/Download
MyAppInstaller_web_Chpt5.exe3.13 MBUnknown

OPEN

NoneView/Download
MCS_App_chpt7.zip3.49 MBZIP

OPEN

NoneView/Download
MyAppInstaller_web_Chpt6.exe7.81 MBUnknown

OPEN

NoneView/Download

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.