基于EA、LMI与ANN的平流层飞艇纵向智能增稳控制系统设计  

Stratospheric Airship Longitudinal Intelligence Stability Augment System Design Based on EA,LMI and ANN

作　　者：王润平[1] 安锦文[2] 吴梅[2] 陈澜[2] 

机构地区：[1]95948部队 [2]西北工业大学自动化学院

出　　处：《弹箭与制导学报》2007年第2期21-25,28,共6页Journal of Projectiles,Rockets,Missiles and Guidance

摘　　要：文中描述分析了平流层飞艇的特点,建立了飞艇的六自由度数学模型,并对其进行了线性化,获得了状态空间表达式,对得到的线性系统进行了特性分析,并将纵向运动划分为三个模态。为了改善飞艇的稳定性,在空域中选定若干设计点,利用基于LMI的EA对其进行了控制器设计,然后利用神经网络形成了神经网络控制器,达到了控制器全空域泛化的目的。结合实际数据对加入控制器后的飞艇模型进行仿真,仿真结果表明了这种设计方法的有效性。This paper described and analyzed the characteristics of stratospheric airship, presented stratospheric airship＇s six-degree-of-freedom nonlinear dynamic model, and gained the state-space expression by linearizing the nonlinear model. We analyzed the linear matrix inequality （LMI） system and divided into three modes for the longitudinal motion. For the improvement of airship＇s stability, we selected some design point in all the sky, and designed the controller at these point by eigenstructure assignment （EA） based on LMI. Then we formed artificial neural network（ANN） controller via ANN for the objective that can extend this controller into all the sky. Combined the actual data, we did the simulation for the airship＇s nonlinear model added ANN-controller. The simulation results showed the validity of this design method.

关 键 词：平流层飞艇 动力学模型 线性矩阵不等式 特征结构配置 神经网络 

分 类 号：V274[航空宇航科学与技术—飞行器设计]