摘要
东江惠州段水质直接影响着香港和深圳的淡水供应质量 ,本文根据东江水质自动监测系统的分布情况 ,提出了由上游水质预测下游水质和当前水质预测未来水质的两种基于自适应神经网络的东江惠州段水质预测建模方法 ,给出了基于误差梯度信息的离散神经网络自适应学习算法 ,用李亚普诺夫方法证明了该自适应神经网络算法的收敛性 ,仿真结果证明了该方法具有较高的预测精度 ,且方法简便、适用对象广泛。
The fresh water quality feed to Hong Kong and Shenzhen is affected directly by the water quality of Huizhou reach of the East River. According as the distribution of automonitors, two adaptive neural network based prediction modelling methods of water quality for a river reach are put forward. One is that anticipating the lower course water quality by measuring the upriver water quality. Another is that estimating the future state with current water quality in a same position. The discrete adaptive learning algorithms based on grads information is provided. The adaptive NN algorithm is proved convergent by Lyapunov method. Simulation results prove that the proposed approaches have high precision,good adaptability and extensive applicability.
出处
《系统工程》
CSCD
北大核心
2001年第1期89-93,共5页
Systems Engineering
基金
国家自然科学基金! ( 69874 0 0 5)
广东省环保局资助项目! ( 980 0 1 0 )
广东省工业自动化重点学科资助项目
关键词
自适应神经网络
水质预测
河流
建模
环境问题
水污染
Adaptive Neural Networks
Grads Information
Water Quality Prediction
The East River
