摘要
阐述了洪水预报BP网络模型的构造和数据处理方法,在对BP网络的学习率和动量项的自适应调整算法进行改进的基础上,建立了降雨水位预报BP网络模型,实现对洪水水位的直接预报;同时还提出以点雨量作为模型输入的建模方案,有效地改善了BP网络的模拟性能.
In this paper, the adaptive adjustment algorithm of the momentum and learning rate of BP Network is improved. The structure of BP network and the data processing method of flood forecast model are discussed. In order to predict the flood water level directly, the rainfaU-water level forecast model is built. Furthermore, a new modeling scheme, in which the point rainfall data are used as the model input, is proposed, and the accuracy of BP network is improved.
出处
《中国科学院研究生院学报》
CAS
CSCD
2007年第1期81-85,共5页
Journal of the Graduate School of the Chinese Academy of Sciences
基金
国家自然科学基金项目(40571117)
中国科学院知识创新工程重要方向项目(KZCX3-SW-338)
中意合作项目"洪水风险规划
监测和预报集成系统"资助
关键词
BP网络
降雨-水位模拟
点雨量
BP network, rainfall-water level modeling, point rainfall
