摘要
提出了一种多层前馈神经元网络训练算法．该算法具有指数收敛速度、能克服病态、简单易用．并给出一种计算水电站下游的不稳定流的神经元网络方法，实例证明其有效性．
This paper presents a new multilayer feedforward neural networks learning algorithm. It is of exponential convergent speed,and also applicable for the ill-conditioned situations. Its convergence is discussed too The practical resuits indicate effectiveness of the algorithm.
出处
《水电能源科学》
北大核心
1995年第3期173-178,共6页
Water Resources and Power
基金
"八五"项目
关键词
水电站
不稳定流
神经元网络法
计算
multilayer feedfporward networks, Learning algorithm, Exponential convergence speed, unsteady flow.
