弹性高超声速飞行器抗输入饱和动态神经网络控制

针对弹性高超声速飞行器纵向动力学模型,分析了弹性模态的动态过程,研究了控制器设计问题。控制器设计采用内环与外环相结合的控制方式,内环控制器设计是以攻角和俯仰角速度为被控对象,将俯仰舵偏角作为控制输入,在考虑弹性模态动态过程的情况下设计抗输入饱和辅助系统处理输入饱和问题,采用反演设计方法设计控制器,并利用全局调节动态神经网络逼近控制输入的饱和特性;外环控制器设计是以飞行速度为被控对象,将燃料当量比作为控制量,保证弹性模态良好动态特性的情况下,设计抗饱和辅助系统处理输入饱和问题,采用终端滑模控制方法设计控制器,利用全局调节动态神经网络逼近输入饱和特性,理论分析证明了闭环系统所有信号均有界并指数收敛到原点的一个领域内,仿真分析验证了设计的控制策略可以保证弹性模态具有良好动态特性的情况下,系统的状态可以有效跟踪期望指令信号。

Anti-input windup dynamic neural networks control for flexible hypersonic vehicles

The dynamic process of the flexibility mode is analyzed and the controller is designed based on the inner loop and outer loop systems for the flexible hypersonic vehicle. Considering the angle of attack and the pitch rate as the states of the inner loop systems and the elevator deflection as the control input, the controller is designed by the backstepping method, the question of control input windup is solved with the anti-input windup supplementary systems in case of considering the dynamic process of the flexibility mode. The windup character of control input is approximated by fully tuned dynamic neural networks. Considering the flight speed as the states of outer loop systems and the equivalence ratio as the control input, the controller is designed based on the terminal sliding mode control method. The question of control input windup is solved with the anti-input windup supplementary systems and assuring the favorable dynamic character of flexibility mode, the input windup character is approximated by the fully tuned dynamic neural networks. Based on the stability theorem, all the signals of the systems are bounded and exponentially converge to the neighborhood of the origin globally. The simulation results demonstrate that order signals can be traced effectively by states of the system and the dynamic character of the flexibility mode is favorable.