摘要
在船舶初始设计阶段，船舶的主尺度和船型系数可基本确定，这时如能建立这些参数与船舶操纵性之间的关系，就能在初始设计阶段预报船舶的操纵性能．通常，采用线性回归方法建立这些关系，但其预报结果精度难以保证．本文基于人工神经网络理论，提出了新的预报操纵性能的方法。
At the preliminary stage of ship design, building up the relationships among the ship maneuverability and some main dimensions and form coefficients for quickly predicting the ship maneuverability are important things. The usual way to do this is by using the regression analysis, but some problems are often encountered .This paper puts forward a new method based on the neural network theory to quickly predict the ship maneuverability. The calculating results are presented for a fleet.
出处
《武汉交通科技大学学报》
1997年第1期1-5,共5页
Journal of Wuhan University of Technology(Transportation Science & Engineering)
基金
国家青年自然科学基金
关键词
船舶操纵性
神经网络
预报
ship maneuverability
artificial neural networks
B P algorithm
