人工智能驱动的可持续环境基础研究系统  

AI-driven sustainable environmental foundation research system

作　　者：李佳礼 刘杰 胡承志[2,3] 曲久辉 LI Jiali;LIU Jie;HU Chengzhi;QU Jiuhui(Department of Chemistry,National University of Singapore,Singapore 117543,Singapore;State Key Laboratory of Environmental Aquatic Chemistry,Research Center for Eco-Environmental Sciences,Chinese Academy of Sciences,Beijing 100085,China;University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]新加坡国立大学化学系,新加坡117543 [2]中国科学院生态环境研究中心环境水质学国家重点实验室,北京100085 [3]中国科学院大学,北京100049

出　　处：《环境工程学报》2024年第5期1205-1218,共14页Chinese Journal of Environmental Engineering

基　　金：国家自然科学基金基础科学中心项目(52388101)。

摘　　要：近年来,复合环境问题多发,应用传统研究方法难以灵活快速响应新型环境挑战。人工智能(AI)技术的快速发展为可持续环境基础研究提供了新思路。本文首先提出了复杂变化的环境物质系统的关键信息近似方法和系统方程,基于此提出了稳态及动态环境物质系统的机器学习框架。文章深入探讨了人工智能在分子性能预测、材料设计、蛋白质合成等基础学科领域的应用,并阐述了这些技术在新污染物预测、环境功能材料设计、活性污泥菌群优化等环境研究领域的迁移应用。基于此,详细介绍了人工智能在水处理工艺优化和机理研究中的应用,展示其解决环境问题的应用潜力,并提出了环境系统多组分信息融合策略。然而,要实现人工智能技术在环境领域的广泛应用,仍需克服系统关键描述信息构建难、数据匮乏和多组分信息融合技术的优化等挑战。In recent years,complex environmental problems have become increasingly common,and traditional research methods struggle to respond flexibly and swiftly to new environmental challenges.The rapid development of artificial intelligence(AI)technology offers fresh perspectives for sustainable environmental foundation research.This paper first introduces key information approximation methods and system equations for complex dynamic environmental substance systems,based on which a machine learning framework for steady-state and dynamic environmental substance systems is proposed.The paper delves into the application of artificial intelligence in fundamental scientific fields such as molecule property prediction,material design,and protein synthesis,and further discusses its transfer applications in environmental research,including new pollutants prediction,environmental functional material design,and active sludge microbial community optimization.Based on this,the paper details the application of artificial intelligence in water treatment process optimization and mechanism research,demonstrating its potential in solving environmental issues,and proposes a multi-component information fusion strategy for environmental systems.Nonetheless,realizing the widespread application of AI in the environmental field requires overcoming challenges related to the construction of essential descriptions of the system,data scarcity,and the optimization of multi-component information fusion techniques.

关 键 词：人工智能 环境系统工程 环境应用 水处理 基于物理建模 

分 类 号：X-1[环境科学与工程]