论文部分内容阅读
针对近水面机动情况下潜器的深度及纵倾控制问题提出一种不基于模型的非线性自适应控制策略。其中,一种称为模糊FCMAC的特殊神经网络被用于补偿潜器动态模型的非线性部分。基于李雅普诺夫原理而推导出的在线学习算法用于更新FCMAC的权值。仿真结果表明此控制策略能较好地适应潜器质量、航速及海浪变化,在较大的工况变化范围内保持良好的控制性能。
A non-model-based non-linear adaptive control strategy is proposed for the problem of submarine depth and trim control in the near-surface maneuvering. Among them, a special neural network called fuzzy FCMAC is used to compensate for the nonlinear part of the dynamic model of submarine. The online learning algorithm based on Lyapunov’s principle is used to update the weight of FCMAC. The simulation results show that this control strategy can better adapt to the submarine quality, speed and wave changes, and maintain good control performance in a wide range of working conditions.