摘要
以多操纵面布局飞机为对象,针对其操纵面冗余造成的激励信号线性相关及控制过程中存在的信号激励不足问题,对最小二乘辨识算法进行了改进。新方法将先验经验引入在线辨识过程,可以克服弱信号输入造成的辨识不准确;采用激励-补偿机制,增加激励信号,改善了辨识结果的精度。以六自由度非线性飞机模型为对象进行的姿态控制系统仿真验证表明,当操纵面完好时,该方法能够在线修正建模误差;当飞机操纵面损伤时,该方法能够迅速反映操纵面气动参数的变化,提高了控制系统的适应性和鲁棒性。
The lack of signal strength or collinear signals lead significant bias to the identification result, which are more serious to the new multi-actuator aircraft. An improved least square method is put forward. Prior informations are used to avoid the bad identification result. Injection signals are used to break the collinear relations among the signals, at the same time, that does not affect the rigid body motion of the aircraft. The simulations show that the identify results can follow the parameters changed rapidly which are led by modeling error or the damage of the aircraft actuator. The adaptability and the robustness of the control system are improved.
出处
《飞行力学》
CSCD
北大核心
2008年第1期5-9,共5页
Flight Dynamics
基金
国家973基金资助项目(5132004)
关键词
在线参数辨识
最小二乘法
多操纵面
on-line parameter identification
least square method
multi-actuator
