基于时间序列、主成分聚类和机器学习耦合作用的水环境质量研究:全过程视野下的解析  

Research on Water Environmental Quality based on Space-time Evolution,Principal Component Clustering and Machine Learning Coupling:Perspective from A Whole-process

作　　者：刘瑶[1,3] 胡静 练小可 欧阳吴渝 沈滟奇 LIU Yao;HU Jing;LIAN Xiaoke;OUYANG Wuyu;SHEN Yanqi(Key Laboratory of Eco-Environment of Three Gorges Region,Ministry of Education,Chongqing University,Chongqing 400044,China;National Inland Waterway Engineering and Technology Research Center,Chongqing Jiaotong University,Chongqing 400074,China;ChongQing Architectural Design Institute Co.,LTD.,Chongqing 400015,China)

机构地区：[1]三峡库区生态环境教育部重点实验室,重庆大学,重庆400044 [2]国家内河航道整治工程技术研究中心,重庆交通大学,重庆400074 [3]重庆市设计院有限公司,重庆400015

出　　处：《环境影响评价》2023年第4期116-124,共9页Environmental Impact Assessment

摘　　要：人类活动不断增强导致水环境恶化的问题日益凸显,直接影响着人们的生产生活,因此水环境质量受到密切关注。但现有对水环境问题的研究多是基于单一维度的阐述,缺乏系统性解析。针对上述问题,该文结合近年来快速发展的智能化、大数据技术,通过时间序列、机器学习和主成分聚类等方法,建立了一套涵盖数据抓取与整理、水环境时空演变分析、敏感因子路径辨识和管理策略整合的全过程链模型。基于上述方法体系,以长江流域为对象,对流域水环境质量开展溯源分析。As the deterioration of water environment caused by human activities has become increasingly prominent,which directly affects our production and life.Therefore,the quality of water environment has become the focus in the environment research field.However,most of the existing studies are based on a single-dimension without systematic analysis.According to the above problem,this study combined the rapid development in recent years with intelligentization and big data technology.And through the time series,space-time evolution coupled analysis of machine learning and the methods of principal component cluster,established a model covering the whole process,which organizes the space-time evolution of water environment analysis,and distinguishes the sensitive-factor in its data.Based on the above method system,we can analyze the river basin water environment quality.

关 键 词：水环境质量 时空演变 主成分分析 机器学习 

分 类 号：X143[环境科学与工程—环境科学]