基于RBF神经网络的减摇鳍自适应滑模控制  

Adaptive Sliding Mode Control for Ffin Stabilizer Based on RBF Neural Network

作　　者：石为人[1] 陶芬[1] 张元涛[1] 

机构地区：[1]重庆大学自动化学院,重庆400030

出　　处：《控制工程》2012年第6期978-981,986,共5页Control Engineering of China

基　　金：重庆市科技攻关项目(CSTC2007AB3035;CSTC2008AB3049)

摘　　要：为减轻船舶在大风浪中剧烈的横摇,减摇鳍是目前应用最广泛的减摇装置。针对船舶减摇鳍系统的非线性和不确定性,在系统不确定性函数结构未知的情况下,提出一种RBF神经网络自适应滑模控制方法。采用RBF神经网络逼近系统不确定动态,并设计权值的自适应律,结合滑模控制增强系统的鲁棒性。在不同有义波高和不同浪向角下,建立随机海浪的干扰模型,应用simulink对系统进行仿真。仿真结果表明,该控制策略在各种海况下,均具有良好的减摇效果和较强的鲁棒性。The ship fin stabilizer is a typical uncertain nonlinear system, and it is often assumed that the system uncertain function is unknown. A basic RBF neural network adaptive sliding mode control （SMC） is proposed to reduce the violent roll of the ship in rough seas. It uses RBF neural network to approximate the system uncertain dynamics and design the weights adaptive laws, and it utilizes SMC to guarantee the system robustness. Furthermore, Simulink is used to simulate the system at various sea conditions of different sig- nificant wave height and wave direction angles. T-he simulation results show that the improved RBF neural network adaptive sliding mode control is effective in the control of the fin stabilizer system and has strong robustness.

关 键 词：减摇鳍 RBF神经网络 自适应控制 滑模控制 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]