基于支持向量机的白云湖水质评价

Evaluation of water quality of Baiyun Lake based on support vector machine

利用支持向量机模型良好的推广和泛化能力,以及在处理分类问题及小样本问题方面的优势,构建了湖泊水质评价模型,并运用此模型对白云湖的水质评价进行了研究。于2011年对广州市白云湖进水口A和出水口E以及湖区内B、C、D共5点分别进行了3次水质监测(1月份、4月份和8月份)。分析结果表明,A、B、C、D、E 5点1月份的水质除B点为Ⅳ类水外其余4点均为Ⅴ类水,4月份除A点为Ⅴ类水外其余4点均为Ⅳ类水,8月份除A为Ⅳ类水外其余4点均为Ⅱ类水,白云湖在经过开始的不稳定状态后,正在逐渐实现其净化水质的设计作用。相对于常规的评价方法,所得结果更为科学、合理。

Support vector machine( SVM) is easy to be spread and generalized and has an advantage in dealing with classification and small sample size problems. An evaluation model of lake water quality was proposed based on SVM and was used to evaluate the water quality of Baiyun Lake in Guangzhou. The water quality in the intake( A),outlet( E) and central area( B,C,and D) of Baiyun Lake was monitored three times in2011,respectively. The analysis results showed that the water qualities of Baiyun Lake except site B( class IV)were belonging to class V in January. In April,except site A( class V),the remaining sites were all classified as grade IV. In August,the water quality of site A was transferred to class IV and the other sites were improved to class II. After the unstable state,Baiyun Lake is gradually achieving the design goal of water purification. Compared with the results evaluated by other common methods,these results were more scientific and reasonable.