关键词: 地下水水质评价, 主成分分析法, 秩和比法, 降维, 多元指标

Abstract: The groundwater quality is affected by many indicators, and information between each indicator is often overlapped and interrelated, which is an extremely complex integrated system. Currently, most of the water quality evaluation methods take no or insufficient overlapped information between each indicators into consideration in the evaluation process, which makes the results biased. To solve these problems, this paper took nine water quality evaluation indicators such as nitrite, nitrate and total hardness from the limestone aquifer in the Xiandewang coal mine, and extracted 3 water quality components based on principal component analysis(PCA). The weight coefficient of each principal component was obtained by using the rank sum ratio method(RSR), and the groundwater quality evaluation model was established by combining GIS technology, and the aquifer water quality was analyzed. The results confirmed the feasibility and superiority of the PCA-RSR model in groundwater quality evaluation, and showed that the water quality was mainly in class Ⅲ, class Ⅳ and class Ⅴ. The main pollution indicators were total hardness, sulfate, ammonia nitrogen, nitrite and so on, which provided important reference for groundwater quality remediation in the study area.

Key words: groundwater quality evaluation, principal component analysis, rank sum ratio method, dimensionality reduction, multiple indicators