| 二阶对角递归神经网络船舶横摇运动预测 |
| |
| 引用本文： | 李占英,王科俊,张明君,唐墨.二阶对角递归神经网络船舶横摇运动预测[J].华中科技大学学报(自然科学版),2011(6):125-128. |
| |
| 作者姓名： | 李占英 王科俊 张明君 唐墨 |
| |
| 作者单位： | 哈尔滨工程大学自动化学院;大连理工大学城市学院电子与自动化学院; |
| |
| 基金项目： | 国家自然科学基金资助项目(60975022); 国家高技术研究发展计划资助项目(2008AA01Z148) |
| |
| 摘 要： | 提出一种船舶横摇时间序列预测方法.该方法使用在隐层具有2个反馈权值的对角递归神经网络进行预测,给出了此网络易于实现的动量梯度学习算法(DBP),并对其收敛性进行了验证.运用该模型对我国某型船舶在横浪中航行情况进行预测,结果表明:本网络可以储存更多的历史数据,有更好的记忆性能,所使用的模型比DRNN模型及前向网络BP模型能快速、准确地预测船舶横摇运动时间序列,仿真实验验证了该方法的可行性与有效性.
|
| 关 键 词： | 船舶横摇运动 对角递归神经网络 动量梯度学习算法 时间序列预测 非线性 前向神经网络 |
Second-order diagonal recurrent neural network approach to ship roll prediction |
| |
| Li Zhanying, Wang Kejun Zhang Mingjun Tang Mo.Second-order diagonal recurrent neural network approach to ship roll prediction[J].JOURNAL OF HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY.NATURE SCIENCE,2011(6):125-128. |
| |
| Authors: | Li Zhanying Wang Kejun Zhang Mingjun Tang Mo |
| |
| Institution: | Li Zhanying1,2 Wang Kejun1 Zhang Mingjun2 Tang Mo1(1 College of Automation,Harbin Engineering University,Harbin 150001,China,2 School of Electronic Engineering and Automation,City Institute,Dalian University of Technology,Dalian 116600,China) |
| |
| Abstract: | A method to predict ship rolling time series was proposed.A diagonal recurrent neural network in the hidden layer with two recurrent weights was used to predict time series.A generalized dynamic back-propagation(DBP) algorithm was used to train;moreover,convergence of DBP was derived.Using this model,the situation of one certain type of ship sailing in the beam sea condition was predicted.Simulation results show that the network can store more historical data,and possess a better memory performance.The pres... |
| |
| Keywords: | ship rolling motion diagonal recurrent neural network dynamic back-propagation time series prediction nonlinear feed-forward neural network |
| 本文献已被 CNKI 等数据库收录！ |
