基于两种方法的石佛寺人工湿地水质综合评价与分析  

Comprehensive Evaluation of Water Quality and Analysis Based on Two Methods in Shifosi Constructed Wetland

作　　者：徐悦[1] 周林飞[1] 成遣[1] 

机构地区：[1]沈阳农业大学水利学院,沈阳110866

出　　处：《节水灌溉》2017年第4期106-109,117,共5页Water Saving Irrigation

基　　金：辽宁省科学事业公益研究基金(2015003006);辽宁省财政科研基金(14C030)

摘　　要：采用熵权赋权贝叶斯法和BP神经网络模型法两种方法,评价石佛寺人工湿地2009~2014年间汛期和非汛期入口、中间和出口的水质级别。经比较,两种评价结果基本相同,均可用于水质评价,但熵权赋权贝叶斯法计算更简便。水质评价结果表明:出库水质基本都优于入库水质,说明石佛寺人工湿地对水质具有良好的净化效果;非汛期湿地出口水质较汛期差;出库水质均为Ⅲ和Ⅱ,水质状况良好,不存在用水障碍。In this paper,the water quality of inlet, middle and outlet in flood and non-flood period of Shifosi Constructed Wetland in the six years（2009-2014） are evaluated based on the two methods of BP neural network model and Bayesian Method with Entropy Based-Weight. Through comparison, it is found that the water quality evaluation results by the two methods are basically the same and all can be used for water quality evaluation, while Bayesian Method with Entropy Based-Weight is more simple and convenient. The evaluation results show that the export water quality is basically better than inlet water quality, the Shifosi wetland water purification effect is good ; the export water quality of wetland in non-flood season is worse than that in flood season; the export water quality level can reaches II or m , the water quality of Shifosi Constructed Wetland is good, so there is no water obstacle.

关 键 词：BP神经网络 熵权赋权贝叶斯 水质综合评价 分析 

分 类 号：X824[环境科学与工程—环境工程]