节点文献
ARMAX模型与多层BP网络流量演算法比较
Flow Routing Methods Based on ARMAX Model and Multilayer BP Neural Networks
【摘要】 分别给出了将ARMAX模型及多层BP网络应用于流量演算中模型参数确定的方法;提出了一种可进行误差实时校正的改进ARMAX模型,并认为该模型实质上是一种特殊的两层BP网络;用一个实际算例将改进ARMAX模型与多层BP网络两种方法作了对比分析.
【Abstract】 In order to deal with the nonlinear problems in flow routing, two methods based on AR- MAX model and multilayer BP neural networks are developed respectively. An improved ARMAX model is presented which could rectify errors in real-time. This model is a special kind of two-layer BP network actually. The comparison of the efficiency of the two methods is analyzed theoretically. The BP model could simulate the real nonlinear hydrologic systems directly. The simulated results indicate that both of the methods are of high efficiency.
【关键词】 流量演算;
非线性;
多层神经元网络;
自回归滑动平均模型;
【Key words】 flow routing; nonlinear; multilayer neural networks; ARMAX;
【Key words】 flow routing; nonlinear; multilayer neural networks; ARMAX;
【基金】 湖北省自然科学基金(96J014);华中理工大学能源学院教育基金资助项目
- 【文献出处】 华中理工大学学报 , 编辑部邮箱 ,1998年10期
- 【分类号】P333.6
- 【被引频次】3
- 【下载频次】65