基于DRNN神经网络的PD混合控制技术在船舶动力定位系统中的应用  

Application of DRNN-Based PD Hybrid Control Techniques in Ship Dynamic Positioning Systems

作　　者：夏国清 Corbett Dan R 

机构地区：[1]哈尔滨工程大学动力与核能工程学院,黑龙江哈尔滨150001 [2]University of South Australia Adelaide 5095,Australia

出　　处：《中国造船》2006年第1期48-54,共7页Shipbuilding of China

摘　　要：船舶在海上运动是一种复杂的非线性运动,其水动力系数很难精确确定,而海洋环境的随机干扰因素也在不断地发生变化,因此需要研究具有鲁棒性和自适应能力的船舶动力定位控制技术。P ID控制在优化参数的条件下,对于能够建立精确数学模型的确定性系统具有鲁棒性好和可靠性高的特点,但对于船舶运动这样复杂的非线性系统其控制效果不理想,而神经网络具有自学习和自适应能力,因此需要结合两者的特点,设计自适应能力强、鲁棒性好的控制技术。本文研究了基于DRNN神经网络的PD混合控制技术,并将其应用到船舶动力定位系统。仿真结果表明该方法有效,且具有较好的鲁棒性和自适应能力,提高了动力定位系统的精度和性能。The ship motion is a type of complicated nonlinear motions. It is very difficult to get the accurate hydrodynamic parameters when the ship is in sea. Moreover, the external disturbance forces coming from environment （wind, wave and current） are changing at any moment. So it is necessary to study new control techniques with robust and adaptive properties, PID controller has good robust and high precise character for that can get accurate model of definite systems in math. But PID controller is not satisfactory for complicated nonlinear systems like ship motion. The neural network has self-learning and self-adapting ability, so we can combine the PID with neural network to build the strong robust and adaptive controller. In this paper DRNN-based PD hybrid controller is studied. The simulation results show that the method is effective with strong robust and adaptive capability, and it can be applied to marine dynamic positioning systems.

关 键 词：船舶、舰船工程 动力定位系统 数学模型 PD DRNN神经网络 混合控制 

分 类 号：U661.338[交通运输工程—船舶及航道工程] U666.124[交通运输工程—船舶与海洋工程]