【摘要】 以拖曳体的拖曳水池样机试验取得试验数据作为训练样本,采用LMBP算法,建立基于神经网络理论构建的可控制水下拖曳体轨迹与姿态水动力控制数值模型,并进行LMBP模型仿真检验。结果表明,利用所建全的神经网络模型对拖曳体在一定控制动作下的水动力响应预报是令人满意的。更多还原

【Abstract】 A hydrodynamic neural network model for a trajectory and attitude controllable underwatertowed vehicle is established based on LMBP algorithm. The training samples are provided fromtesting da-ta of the towed vehicle prototype towing experiments conducted in a towing tank and the model can thenbe established. The numerical simulation results in the paper indicate that forecast of the hydrodynamicand control performances of the underwater towed vehicle under a given control manipulation by the estab-lished neural network model is satisfactory.更多还原