载人潜水器潜浮运动的模拟

为了预报中国正在研发的深海载人潜水器的潜浮性能,三自由度的动力学模型被用于仿真它在垂直面内的潜浮运动.作用于潜水器上的水动力系攻角的非线性函数,可用拖曳水池试验加以测定.利用这些试验值,GRNN神经网络方法可以辨识出大攻角范围内的水动力函数.本文通过数值计算结果揭示了其实际的应用.

Simulation of Dive Motion of a Deep Manned Submersible

In order to predict the dive motion of a deep manned submersible currently developed in China,a three-degree-of-freedom dynamic model is developed to simulate its dive motion restricted in the vertical plane. Being the nonlinear function of attack angle, the hydrodynamic forces acting upon the submersible are determined by a towing tank test.Based on these experimental values,the hydrodynamic forces are identified by a general regression neural network (GRNN) over a wide range of attack angles.Results of numerical examples are presented in this paper to explain practical applications of the method.