基于改进神经网络算法的舰船操纵性能预报系统

发布时间：2018-05-12 05:21

本文选题：神经网络 + 舰船操纵稳定性　；参考：《舰船科学技术》2017年22期

【摘要】：针对传统舰船操纵性能预报系统预报精度低、稳定性差、效率慢等弊端,设计一种新的基于改进神经网络的舰船操纵性能预报系统。对标准神经网络算法容易陷入局部最优解的不足进行改进,用共轭梯度法取代原梯度下降算法,介绍了改进后的神经网络算法。通过矩阵方式对舰船操纵稳定性的构成进行描述,利用改进的神经网络算法设计舰船操纵稳定性预报系统。实验结果表明,所设计系统预报精度高、稳定性强、效率快,能有效完成舰船操纵稳定性的预报。
[Abstract]:A new ship maneuvering performance prediction system based on improved neural network is designed to overcome the disadvantages of low prediction accuracy, poor stability and low efficiency in the traditional ship maneuvering performance prediction system. The deficiency of standard neural network algorithm which is easy to fall into local optimal solution is improved. The conjugate gradient method is used to replace the original gradient descent algorithm and the improved neural network algorithm is introduced. The structure of ship maneuvering stability is described by matrix method, and the prediction system of ship maneuvering stability is designed by using improved neural network algorithm. The experimental results show that the designed system has the advantages of high prediction accuracy, strong stability and high efficiency, and it can effectively predict the ship maneuvering stability.
【作者单位】：重庆工程学院;
【分类号】：TP183;U674.70

【相似文献】