| 分布式遗传的船舶航向神经网络优化控制 |
| |
| 引用本文： | 邓华,王仁强,胡甚平,缪克银,杨永前.分布式遗传的船舶航向神经网络优化控制[J].上海海事大学学报,2020,41(4):15-19. |
| |
| 作者姓名： | 邓华 王仁强 胡甚平 缪克银 杨永前 |
| |
| 作者单位： | 江苏海事职业技术学院航海技术学院,南京211170;上海海事大学商船学院,上海201306 |
| |
| 基金项目： | 江苏省教育厅高等学校自然科学研究项目（19KJA150005,19KJD58001,18KJB580003）；江苏省青蓝工程（2019）；江苏省高校实验室研究会研究项目（GS2019YB14） |
| |
| 摘 要： | 针对海上风浪环境对船舶航行的干扰，利用遗传神经网络优化算法设计船舶航向控制器。利用分布式遗传算法（distributed genetic algorithm,DGA）并结合模拟退火算法对常规遗传算法(genetic algorithm,GA)进行改进。利用改进的GA对径向基函数(radical basis function,RBF)神经网络进行优化。利用优化的RBF神经网络对系统不确定项进行逼近，并对控制输入进行补偿实现抗饱和控制。利用三阶干扰观测器对外部扰动实时跟踪并反馈到滑模控制器(sliding mode controller,SMC)设计中。借助SMC设计并结合李雅普诺夫稳定性理论推算出船舶运动控制律，实现船舶运动优化控制。通过实验验证了本文设计的控制器性能较现有的模糊PID控制器和神经网络SMC优越，系统达到稳定的时间短，平均超调量小。
|
| 关 键 词： | 船舶运动 优化控制 径向基函数(RBF)神经网络 分布式遗传算法(DGA) 输入饱和 |
| 收稿时间： | 2020/7/4 0:00:00 |
| 修稿时间： | 2020/9/29 0:00:00 |
Ship course neural network optimal control based on distributed genetic algorithm |
| |
| Institution: | Jiangsu Maritime Institute |
| |
| Abstract: | Aiming at the interference of the sea wind and wave environment on ship navigation, a ship course controller is designed by the genetic neural network optimization algorithm. The distributed genetic algorithm (DGA) and the simulated annealing algorithm are used to improve the conventional GA. The improved GA is used to optimize the radical basis function (RBF) neural network. The optimized RBF neural network is used to approximate the uncertain items of the system, and the control input is compensated to realize the anti saturation control. The third order disturbance observer is used to track the external disturbance in real time and feed it back to the design of the sliding mode controller (SMC). With the help of the design of SMC and the Lyapunov stability theory, the ship motion control law is derived to realize the ship motion optimal control. Experiments show that the controller designed in this paper is of better performance than the existing fuzzy PID controller and the neural network SMC, the stability time of the system is shorter, and the average overshoot is smaller. |
| |
| Keywords: | ship motion optimal control radical basis function(RBF) neural network distributed genetic algorithm (DGA) input saturation |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《上海海事大学学报》浏览原始摘要信息 |
| 点击此处可从《上海海事大学学报》下载免费的PDF全文 |
