Modeling and optimizing the performance of PVC/PVB ultrafiltration membranes using supervised learning approaches

Abstract

17Mathematical models plays an important role in performance prediction and 18 optimization of ultrafiltration (UF) membranes fabricated via dry/wet phase inversion 19 in an efficient and economical manner. In this study, a systematic approach, namely, a 20 supervised, learning-based experimental data analytics framework, is developed to 21 model and optimize the flux and rejection rate of Poly (vinyl chloride) (PVC) and 22Polyvinyl butyral (PVB) blend UF membranes. Four supervised learning (SL) 23 approaches, namely, the multiple additive regression tree (MART), the neural 24 verified by experiments. We hope the approach is able to shed light on how to 34 systematically analyzing multi-objective optimization issues for fabrication conditions 35

A low-cost and high-performance thin-film composite forward osmosis membrane based on an SPSU/PVC substrate

“…Moreover, PVC can be dissolved in various industrial solvents 22 . Hence, PVC can be applied for the fabrication of membranes via the nonsolvent-induce phase separation (NIPS) process 20 , 22 – 24 . However, due to its inherent hydrophobicity, hydrophilic modification is necessary for fabricating PVC membranes.…”

Design and development of multilayer cotton masks via machine learning

“…Another study by Chi et al. evaluated the use of four different supervised learning approaches for the modeling of poly(vinyl chloride)/polyvinyl butyral ultrafiltration membranes [ 13 ]. They concluded that the neural network model is preferable which was further confirmed with experimental data.…”