| 基于模糊神经网络的水下潜器多变量解耦控制研究 |
| |
| 引用本文： | 林孝工,付明玉. 基于模糊神经网络的水下潜器多变量解耦控制研究[J]. 哈尔滨工程大学学报, 2003, 24(5): 526-529 |
| |
| 作者姓名： | 林孝工 付明玉 |
| |
| 作者单位： | 哈尔滨工程大学,动力与核能工程学院,黑龙江,哈尔滨,150001 |
| |
| 摘 要： | 潜器在水下进行特殊作业时，需保持空间六自由度的定位姿态，对定位的控制存在叉影响，因而给潜器的精确控制带来困难。针对这一控制非线性系统，提出了一种基于模糊神经网络(DFN)控制方法，该方法对非线性多变量动态系统具有较强的处理能力；它将多输入多输出的复杂系统转变成神经网络设计和耦合补偿结构，预估补偿器方法是通过预定试验的离线学习，获得耦合的预估知识，然后在线完成补偿，仿真结果显示该方法具有很好的鲁棒性。
|
| 关 键 词： | 非线性系统 动态特性 模糊神经网络 |
| 文章编号： | 1006-7043(2003)05-0526-04 |
| 修稿时间： | 2003-01-17 |
Multivariable decoupling control of underwater vehicle based on fuzzy neural network |
| |
| LIN Xiao-gong,FU Ming-yu. Multivariable decoupling control of underwater vehicle based on fuzzy neural network[J]. Journal of Harbin Engineering University, 2003, 24(5): 526-529 |
| |
| Authors: | LIN Xiao-gong FU Ming-yu |
| |
| Abstract: | When an underwater vehicle is carrying out a special task,it must take 6-degree of freedom dynamic positioning posture,which has an intersect-influence on the precision of this control system.For this reason,a control method based on fuzzy neural network (FNN) is introduced for nonlinear system,which does better in dealing with nonlinear dynamical system.The method transforms the complex multi-input /output system into a neural network design and coupling compensation configuration.The pre-compensation is an off-line learning method by predesination test,and completes online compensation by a pre-estimating knowledge of coupling. Simulation results show that the method has very good robustness. |
| |
| Keywords: | not line system dynamic characteristic fuzzy neural networks |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
