摘要
建立基于小波神经网络的预测模型,以不同时间滞差和影响因子组合作为输入变量,对海河流域四个监测断面的溶解氧浓度进行短期预测.结果表明,基于溶解氧历史数据的小波神经网络预测模型精度更高,可用于天然水体的水质预测,为水质管理提供更客观的参考和依据.
The prediction models based on wavelet neural network were established and applied to four certain monitoring sections of Haihe basin. Different combinations of water quality parameters were set as input variables to predict dissolved oxygen. The results demonstrate that models with historical data of the target variable are better fitting with the real data, and have a higher accuracy. Therefore, it will provide significant decision support for water protection and water environment treatment.
出处
《数学的实践与认识》
北大核心
2016年第16期122-127,共6页
Mathematics in Practice and Theory
基金
国家自然科学基金(51478025)
关键词
小波神经网络
水质预测
溶解氧
天然水体
wavelet neural network
water quality forecast
dissolved oxygen
natural water body
