摘要
利用神经网络的非线性拟合能力,建立了基于广义回归神经网络的"船型要素-船体阻力"数学模型,提高了模型的拟合精度。同时结合遗传算法的非线性寻优能力,利用改进的遗传算法完成了船型要素的优化设计。优化结果可以作为玻璃钢渔船初步设计的技术参考。
The mathematical model of ＂ship parameters - ship resistance＂for FRP fishing vessels, based on generalized regression neural networks （GRNN）, is established by using the nonlinear fitting ability of neural networks to improve the model fitting precision. At the meantime, the ship parameters＇ optimizing design is accomplished by the improved genetic algorithm （GA） with its nonlinear optimization ability. The optimized results can be used as the reference for the preliminary design of FRP fishing vessels.
出处
《船舶工程》
CSCD
北大核心
2012年第4期18-20,65,共4页
Ship Engineering
基金
国家公益性行业科研专项(渔业节能关键技术研究与重大装备开发
201003024)
