摘要
建立了镉-铜-阿特拉津三元复合体系中沉积物吸附阿特拉津的BP神经网络模型,模型训练集、验证集、预测集的均方差分别为0.070,0.094和0.110,模型输入数据与其模拟值相关系数为0.974,验证实验表明该三元复合体系中阿特拉津吸附模型预测值与实验值的平均偏差为7.05%(7组数据)。所建模型的预测结果显示,镉-铜-阿特拉津三元复合体系中阿特拉津的主要吸附位是铁氧化物,其次是有机质,锰氧化物对阿特拉津的吸附具有抑制作用。与Cu2+相比,Cd2+的添加对复合体系中阿特拉津的吸附作用影响更显著,且复合污染效应主要发生在锰氧化物和有机质上,并随锰氧化物和有机质含量的不同表现出抑制作用或促进作用。
A BP neural network model was established to simulate the adsorption process of atrazine in the ternary system of cadmium-copper-atrazine.The mean square deviations of the training set,validation set and prediction set of the model were 0.070,0.094 and 0.110,respectively.The correlation coefficient between the input data and the simulated ones was 0.974.The mean deviation between the predicted values by the BP neural network model and the experimental ones was 7.05%（n=7） based on the validation experiments.The predicted results of the atrazine adsorption by the proposed BP model show that in the ternary system of cadmium-copper-atrazine,the main adsorption site of atrazine is the iron oxides,then the organic matters,and the manganese oxides restrain the atrazine adsorption.Compared with Cu2＋,Cd2＋ influences the atrazine adsorption more significantly,and the combined effect occurs mainly on the manganese oxides and the organic matters and exhibits antagonistic or synergistic effect on the atrazine adsorption with different contents of manganese oxides and organic matters.
出处
《水利水电科技进展》
CSCD
北大核心
2011年第1期37-41,共5页
Advances in Science and Technology of Water Resources
基金
国家自然科学基金(50879025)
关键词
阿特拉津
重金属
BP神经网络
沉积物
抑制作用
促进作用
atrazine
heavy metal
BP neural network
sediment
antagonistic effect
synergistic effect
