摘要
针对目前因雨水径流水质数据匮乏而难以对其污染特征进行有效综合评价的问题,以重庆市秀山县城区某区域几种典型下垫面为研究对象,对2017年两场典型降雨(7月8日与8月8日)的径流污染物进行监测,利用自组织映射(self-organizing map,SOM)神经网络聚类算法对雨水径流水质进行综合评价,对各监测点降雨径流污染特征进行主成分分析,确定初期降雨的主要污染物,并对SOM评价结果进行验证。结果表明:各下垫面大致污染程度情况为城市道路>居住区>广场>屋面,主要污染物为总悬浮物(total suspended solid,TSS)与化学耗氧量(chemical oxygen demand,COD)。同时,研究结果证实SOM法是评估雨水径流污染特征的一种有效方法。
It is difficult to effectively evaluate characteristics of the runoff water quality pollution because of lacking data.The runoff pollutants from several typical underlying surfaces in Xiushan Country of Chongqing on 8th July and 8th August of 2017 were analyzed.An algorithm based on self-organizing map(SOM)cluster was used to comprehensively assess the quality of stormwater.Meanwhile,the principal component analysis(PCA)of runoff quality parameters for each monitoring site was carried out and the evaluation results based on SOM were verified by comparing PCA results.The results show that the pollution level of each underlying surface is generally in the order of urban roads>residential area>square>roof,and the main pollutants are total suspended solid(TSS)and chemical oxygen demand(COD).It also demonstrates that the SOM method is effective for runoff pollution evaluation.
作者
甘春娟
朱子奇
刘梦一
熊毅
GAN Chunjuan;ZHU Ziqi;LIU Mengyi;XIONG Yi(Chongqing Municipal Research Institute of Design Ltd.Co.,Chongqing 400012,China;School of River and Ocean Engineering,Chongqing Jiaotong University,Chongqing 400074,China;Chongqing Xiheng Engineering Consulting Ltd.Co.,Chongqing 401123,China)
出处
《三峡生态环境监测》
2020年第4期73-81,共9页
Ecology and Environmental Monitoring of Three Gorges
基金
重庆市“留创计划”资助项目(cx2017065)。
关键词
SOM
主成分分析
水质评价
雨水径流
下垫面
SOM
principal component analysis
water quality assessment
stormwater runoff
underlying surface
