超算与机器学习辅助高渗透性海水反渗透膜系统优化设计  

Supercomputing and machine learning-aided optimal design of high permeability seawater reverse osmosis membrane systems

作　　者：罗玖 李明恒 Eric M.V.Hoek 衡益 Jiu Luo;Mingheng Li;Eric M.V.Hoek;Yi Heng(School of Computer Science and Engineering,Sun Yat-sen University,Guangzhou 510006,China;National Supercomputing Center in Guangzhou(NSCC-GZ),Guangzhou 510006,China;Guangdong Province Key Laboratory of Computational Science,Guangzhou 510006,China;Department of Chemical and Materials Engineering,California State Polytechnic University,Pomona CA 91768,USA;Department of Civil&Environmental Engineering,California NanoSystems Institute and Institute of the Environment&Sustainability,University of California,Los Angeles(UCLA),Los Angeles CA 90095,USA;Energy Storage&Distributed Resources Division,Lawrence Berkeley National Laboratory,Berkeley CA 94720,USA)

机构地区：[1]School of Computer Science and Engineering,Sun Yat-sen University,Guangzhou 510006,China [2]National Supercomputing Center in Guangzhou(NSCC-GZ),Guangzhou 510006,China [3]Guangdong Province Key Laboratory of Computational Science,Guangzhou 510006,China [4]Department of Chemical and Materials Engineering,California State Polytechnic University,Pomona CA 91768,USA [5]Department of Civil&Environmental Engineering,California NanoSystems Institute and Institute of the Environment&Sustainability,University of California,Los Angeles(UCLA),Los Angeles CA 90095,USA [6]Energy Storage&Distributed Resources Division,Lawrence Berkeley National Laboratory,Berkeley CA 94720,USA

出　　处：《Science Bulletin》2023年第4期397-407,M0004,共12页科学通报（英文版）

基　　金：support provided by Key-Area Research and Development Program of Guangdong Province(2021B0101190003);Zhujiang Talent Program of Guangdong Province(2017GC010576);Natural Science Foundation of Guangdong Province,China(2022A1515011514);financial support from the National Science Foundation(2140946);financial support from the UCLA Sustainable LA Grand Challenge;financial support from China Postdoctoral Science Foundation(2022M723674)。

摘　　要：当前,高渗透性反渗透膜材料的研究引起了广泛的关注,然而高渗透导致的浓差极化与膜污染加剧等瓶颈问题限制了高性能膜材料的应用发展.本工作采用机器学习结合超级计算提出了针对先进反渗透膜材料的组件进水隔网(亚毫米级)与系统(米级)的多尺度优化设计新方法.在进料含盐度35,000 ppm,回收率50%典型工况下,对标目前国际先进海水反渗透淡化工艺,本文提出的优化方案能使淡水制备比能耗(1.66 k Wh/m^(3))降低27.5%,所需膜面积减少约37.2%,系统最大浓差极化因子控制在工程允许范围以内(<1.20),可有效缓解高渗透膜系统中膜污染问题,为高性能膜材料精准设计提供理论依据、计算工具和大数据支撑,有重要的应用潜力.本文提出的机器学习结合超算的多尺度设计新研究范式,突破了基于“试错法”的传统单一尺度组件设计限制,高通量并行计算规模可扩展至93,120核以上,较串行算法计算效率提升3000倍以上,可大幅度缩短高性能膜组件的设计周期.Concentration polarization(CP)should limit the energy and cost reducing benefits of high permeability seawater reverse osmosis(swRO)membranes operating at a water flux higher than normal one.Herein,we proposea multiscale optimization framework coupling membrane permeability,feed spacerdesign(sub-millimeter scale)and system design(meter scale)via computational fluid dynamics and system level modeling using advanced supercomputing in conjunction with machine learning.Simulation results suggest energy consumption could be reduced by 27.5%(to 1.66 kWh m^(-3))predominantly through the use of high permeability SWRO membranes(12.2%)and a two-stage design(14.5%).Without optimization,CP approaches 1.52 at the system inlet,whereas the optimized CP is limited to 1.20.This work elucidates the optimized permeability,module design,operating scheme and benefits of high permeability SWRO membranes in seawater desalination.

关 键 词：Reverse osmosis desalination High permeability membrane Multiscale design optimization Multilayer artificial neural networks SUPERCOMPUTING Enhancement of mass transfer 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]