【作者】 任光； 赵卫军； 牛井超；

【Author】 Ren Guang Zhao Weijun Niu Jingchao (Institute of Marine Engineering, Dalian Maritime University, Dalian 116026,China)

【机构】 大连海事大学轮机工程学院；

【摘要】 针对船舶动态具有大惯性、大时滞、非线性等特性,提出了一种基于RBF网络的船舶航向神经网络模糊控制器。该控制器的特点是将RBF神经网络控制器与模糊控制器有机的结合在一起,利用神经网络的非线性映射及学习能力,将抽象的模糊控制规则转化为神经网络的非线性映射．给出了基于模糊控制规则的RBF神经网络控制器的设计方法,讨论了该控制器的特点。以大连远洋运输公司阳澄湖号油轮为被控对象,利用Matlab语言,针对不同的参数和不同的控制器进行了系统仿真研究,给出多种情况的仿真结果．更多还原

【Abstract】 We propose a neuro-fuzzy ship controller based on Radial Basis Function (RBF) neural networks, in order to compensate the characteristics of ship’s dynamic with large delay and inertia and nonlinear. The main feature of propsal controller is to combine fuzzy controller with neural network controller. We utilize the nonlinear mapping and learning ability of neural networks, and transform a set of fuzzy rules into a nonlinear mapping of neural network. We present the design method of the controller based on both RBF neural networks and fuzzy rules, and discuss the characteristics of the controller. The Tank named "Yang Cheng Hu"to belong to Dalian Ocean Transport Company is selected as a controlled ship. MATLAB is used as simulation languange. Simulation studies are conducted with different parameters and different controllers. Simulation results are given in the paper.更多还原