摘要
在水质评价中,表示某一流域水质量情况的指标众多,而不同的指标间往往存在不同程度的相关性。其中有些相关性甚至非常显著,这使它们提供的信息有可能发生重叠。运用主成分分析综合处理可以产生新的指标,而这些新的指标彼此互补相关又能综合反映水质的情况。通过应用主成分分析逐步降维综合处理的特性对辽河水质进行评价,对辽河4个断面的污染程度做了综合评价。然后再应用主成分分析法选取主要污染指标,结合迭代聚类模型对主成分分析得到的结果进行分类。
In water quality evaluation, different index that can express water quality in certain river basin exist different relation. Some relations are outstanding, so the provided information may overlap. The principal component analysis can deal with new index. And these new index which are related to each other can reflect water condition. With the principal component analysis that is characterized of lowering dimension gradually and dealing comprehensively, this paper evaluates the degree of pollution of four sections in Liaohe river. Then it applies the principal component analysis to collect main pollution index, and combine with system cluster model, so the result can be classified by the K- means dusters.
出处
《水利科技与经济》
2006年第10期660-662,共3页
Water Conservancy Science and Technology and Economy
基金
中国博士后科学基金(2005038201)
国家社会科学基金资助项目(04BJY026)
关键词
主成分分析
水质
辽河
principal component analysis
water quality
Liaohe River
