歼击机智能自适应重构控制方法研究  

Intelligent Adaptive Reconfiguration Control for Fighters

作　　者：张敏[1] 殷利平[1] 申晓宁[1] 

机构地区：[1]南京信息工程大学信息与控制学院,江苏南京210044

出　　处：《计算机仿真》2013年第3期64-67,133,共5页Computer Simulation

基　　金：国家自然科学基金青年项目(61104123);江苏省高校自然科学研究计划项目(20100227);高校科研启动经费(20100301)

摘　　要：针对复杂歼击机控制系统,提出一种智能多模型结构的自适应重构控制方法,使得系统可以在不同的运行环境下对结构故障具有一定的控制重构能力。系统为多模型结构,由若干线性模型构成模型集合,并设计相应的自适应权值调整规则,从而使系统能够实时地获得当前最佳控制输入。再通过引入动态神经网络以确保系统的稳定性,同时避免模型切换等引起的噪声干扰。最后,对某型歼击机进行仿真验证,表明所提重构控制方法对系统在多种故障状态下的稳定运行具有显著的效果。For the complex control system of fighters, an intelligent adaptive reconfiguration control method using multiple models was presented in this paper, which makes the system achieve reconfigure control while some structur- al failures occur. The multiple model structure has a model set composed of several linear models, and the system can obtain the best control input at present by the designed corresponding adaptive weight-adjusting regulation. Then a dynamic neural network was introduced to ensure the whole system being stable and reduce the influence caused by frequent switching. Finally, the simulation was carred out for an uncertain fighter plane, and the result shows that the control method presented is effective for the system＇ s stable working under condition of several failures.

关 键 词：多模型 动态神经网络 重构控制 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]