基于SVM的洋河张家口段水质评价

Water Quality Evaluation of Yanghe River Zhangjiakou Section Base on SVM

将支持向量机(SVM)模式识别功能应用于水质评价过程中,排除主观臆断或数据噪声影响。根据水质评价标准建立SVM分类模型,对洋河张家口16个水质样点进行评价。结果表明,SVM模型精度高,适应性良好;区域水体DO、COD含量为5.36、9.28 g/kg,NH、TP、TN依次为0.083、0.34、2.75 mg/L,分属三、四、三、四、二级水平;区域整体水质属于三等。

In view of the fact that the traditional water quality evaluation is distorted by subjective assumptions or data noise, a water quality diagnosis model of support vector machine （SVM） was proposed, and 16 water quality samples of Yanghe River were evaluated. The results showed that RF model with high accuracy and good adaptability; the contents of DO and COD in the regional water were 5.36 and 9.28 g/kg, respectively, and the contents of NH, TP and TN were 0. 083, 0.34 and 2.75 mg,/L; the overall water quality of the area belongs to III and so on.