大伙房水库水质的BP神经网络模拟预测

Mock prediction of BP neural network for water quality in Dahuofang Reservoir

本文选取大伙房水库2011—2015年的水质指标溶解氧(DO)、高锰酸盐(CODMn)、生化需氧量(BOD5)、氨氮(NH3-N)、总氮(TN)、总磷(TP)6项指标的实测数据作为预测样本,建立了BP神经网络模型,并运用该模型对大伙房水库2016年水质指标进行了验证,结果表明该模型能准确地预测水质指标,可以运用到大伙房水域水质指标的预测和预警预报系统当中。

In the paper,measured data of water quality indexes of Dahuofang Reservoir from 2011 to 2015,including 6 index are selected such as dissolved oxygen(DO),potassium permanganate(CODMn),biochemical oxygen demand(BOD5),ammonia nitrogen(NH3 N),total nitrogen(TN)and total phosphorus(TP).BP neural network model is established.The model is utilized for verifying water quality indicators of Dahuofang Reservoir in 2016.Results show that the model predict the water quality indicators,So that this model can be applied as predictor of water quality as well as early warning forecast system in Dahuofang Reservoir.