Millionfold accelerated AI solver for 3D multi-physical simulations of ultrapermeable membranes

Abstract Solving three-dimensional (3D) multi-physics forward and inverse problems is indispensable for fundamental understanding and optimal design of membrane-based desalination systems. Unfortunately, it is computationally expensive when applying traditional numerical methods. Herein, a modified...

Full description

Saved in:
Bibliographic Details
Main Authors: Yanjin Liu, Jiu Luo, Mingming Huang, Hong Liu, Zhiwei Wang, Yi Heng
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:npj Clean Water
Online Access:https://doi.org/10.1038/s41545-025-00491-1
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849769680732946432
author Yanjin Liu
Jiu Luo
Mingming Huang
Hong Liu
Zhiwei Wang
Yi Heng
author_facet Yanjin Liu
Jiu Luo
Mingming Huang
Hong Liu
Zhiwei Wang
Yi Heng
author_sort Yanjin Liu
collection DOAJ
description Abstract Solving three-dimensional (3D) multi-physics forward and inverse problems is indispensable for fundamental understanding and optimal design of membrane-based desalination systems. Unfortunately, it is computationally expensive when applying traditional numerical methods. Herein, a modified Fourier neural operator (FNO)-based method is proposed to efficiently solve complex 3D multi-physics problems. The intelligent solver solves the 3D forward problems in seconds, which is approximately 105-106 times faster than traditional finite-element based method with a comparable solution quality. The average prediction accuracy is more than 96%. Moreover, the proposed FNO-based method is mesh-independent and has zero-shot super-resolution ability. It can be used to provide a fast solution for the optimal design of membrane module to mitigate concentration polarization and membrane fouling for next-generation ultrapermeable membrane system.
format Article
id doaj-art-df7f5b7bf185451fadf1802c8230260b
institution DOAJ
issn 2059-7037
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series npj Clean Water
spelling doaj-art-df7f5b7bf185451fadf1802c8230260b2025-08-20T03:03:20ZengNature Portfolionpj Clean Water2059-70372025-07-018111010.1038/s41545-025-00491-1Millionfold accelerated AI solver for 3D multi-physical simulations of ultrapermeable membranesYanjin Liu0Jiu Luo1Mingming Huang2Hong Liu3Zhiwei Wang4Yi Heng5School of Computer Science and Engineering, Sun Yat-sen UniversitySchool of Future Science and Engineering, Soochow UniversitySouthern Marine Science and Engineering Guangdong Laboratory (Zhuhai)Chongqing Institute of Green and Intelligent Technology, Chinese Academy of SciencesState Key Laboratory of Pollution Control and Resource Reuse, Shanghai Institute of Pollution Control and Ecological Security, College of Environmental Science and Engineering, Tongji UniversitySchool of Computer Science and Engineering, Sun Yat-sen UniversityAbstract Solving three-dimensional (3D) multi-physics forward and inverse problems is indispensable for fundamental understanding and optimal design of membrane-based desalination systems. Unfortunately, it is computationally expensive when applying traditional numerical methods. Herein, a modified Fourier neural operator (FNO)-based method is proposed to efficiently solve complex 3D multi-physics problems. The intelligent solver solves the 3D forward problems in seconds, which is approximately 105-106 times faster than traditional finite-element based method with a comparable solution quality. The average prediction accuracy is more than 96%. Moreover, the proposed FNO-based method is mesh-independent and has zero-shot super-resolution ability. It can be used to provide a fast solution for the optimal design of membrane module to mitigate concentration polarization and membrane fouling for next-generation ultrapermeable membrane system.https://doi.org/10.1038/s41545-025-00491-1
spellingShingle Yanjin Liu
Jiu Luo
Mingming Huang
Hong Liu
Zhiwei Wang
Yi Heng
Millionfold accelerated AI solver for 3D multi-physical simulations of ultrapermeable membranes
npj Clean Water
title Millionfold accelerated AI solver for 3D multi-physical simulations of ultrapermeable membranes
title_full Millionfold accelerated AI solver for 3D multi-physical simulations of ultrapermeable membranes
title_fullStr Millionfold accelerated AI solver for 3D multi-physical simulations of ultrapermeable membranes
title_full_unstemmed Millionfold accelerated AI solver for 3D multi-physical simulations of ultrapermeable membranes
title_short Millionfold accelerated AI solver for 3D multi-physical simulations of ultrapermeable membranes
title_sort millionfold accelerated ai solver for 3d multi physical simulations of ultrapermeable membranes
url https://doi.org/10.1038/s41545-025-00491-1
work_keys_str_mv AT yanjinliu millionfoldacceleratedaisolverfor3dmultiphysicalsimulationsofultrapermeablemembranes
AT jiuluo millionfoldacceleratedaisolverfor3dmultiphysicalsimulationsofultrapermeablemembranes
AT mingminghuang millionfoldacceleratedaisolverfor3dmultiphysicalsimulationsofultrapermeablemembranes
AT hongliu millionfoldacceleratedaisolverfor3dmultiphysicalsimulationsofultrapermeablemembranes
AT zhiweiwang millionfoldacceleratedaisolverfor3dmultiphysicalsimulationsofultrapermeablemembranes
AT yiheng millionfoldacceleratedaisolverfor3dmultiphysicalsimulationsofultrapermeablemembranes