【摘要】 利用黄河兰州段水质指标浓度时间序列作为学习样本 ,选取了 DO、COD、氨氮、重金属离子等 1 0项指标作为输出参数 ,运用人工神经网络 BP模型中的 Levenberg-Marguardt优化算法对学习样本优化建模 .将优化好的网络的预测结果与实测结果进行了比较 ,表明 BP网络可以很好地用于黄河水质指标值及水质类别的预测更多还原

【Abstract】 Concentration of water qualitative indexes oftime series in Lanzhou reaches of the Yellow Riv- er acted as studying samples.Ten indexes including DO,COD,NH3 - N,heavy metal ions and so on were target parameters.Studying samples were modeled and optimized with Levenberg- Marguardt algorithm of BP neutral network models.Comparison between the predicted results of the network and the deter- mined results indicated thatthe BP network can be satisfactorily used to predictwater qualitative indexes and classification of the Yellow River.更多还原