太湖水质站网数据的主成分分析应用

本文采用主成分分析方法,对太湖湖区的水质站网数据进行了定量化研究,挖掘出对于表征水质状况的主成分变量贡献较大的水质指标,并分析了影响水质的主要原因和各个主成分的环境意义,最终求得主成分的综合得分并进行分级排序,结果表明:前四个主成分已经可以综合90%以上的因子信息,完全可以取代11个指标因子进行综合分析太湖湖区的水质状况。说明主成分分析方法克服了多维指标难以综合分析的缺陷,可能是未来水质综合分析的一种重要方法之一。

Case Study of Application of Principal Component Analysis for Comprehensive Analysis of Water Quality in the Taihu Lake

This article uses the principal components analysis method to carry on the quantitative analysis to the water quality data of Tai Lake station,had discovered the water quality indexes which act on principal components variable in token of water quality condition,and has analyzed the primary causes of the effects on water quality and each principal component environmental significance,finally obtains the comprehensive score calculated by the every principal score and sorted.The results show that the top four principal components,which are extracted from original factor information and can account over 90 % of the total variance,can substitute 11 index factors completely to analyze water quality in the Taihu lake comprehensively.The results also turn out that principal component analysis method has overcome the difficulty of the analysis by synthesis multi-dimensional indicators,and possibly will be one important method of water quality comprehensive analysis in the future.