摘要
利用神经网络网络的自学习能力强、并行计算快、容错性高的特点来评估船舶航行安全性状态,同时利用微粒群算法(PSO)优化反向传播神经网络的连接权值和阈值。阐述反向传播神经网络在船舶航行安全性系统状态评测中的流程,相对于目前的船舶航行安全性预测方法,本文提出的智能模型具有结构简单及预测精度高等优点。
In this paper,the use of self-learning ability of neural network network is strong,fast parallel computing,high fault tolerance features to assess the safety of navigation status,while taking advantage of Particle Swarm Optimization( PSO) to optimize back-propagation neural network( backpropagation artificial neural network) connection weights and thresholds of the network. And elaborated on back-propagation neural network processes navigation safety evaluation of the system status,as opposed to the current method of predicting the safety of ship navigation,intelligent model proposed by the invention has a simple structure,the prediction precision.
出处
《舰船科学技术》
北大核心
2016年第6X期55-57,共3页
Ship Science and Technology
基金
国家自然科学基金资助项目(71361008)
海南省重点科技基金资助项目(ZDXM20130080)
关键词
船舶航行安全
反向神经网络
微粒群算法
safety of navigation
reverse neural network
particle swarm optimization