论文部分内容阅读
针对高超声速飞行器高度非线性及强耦合的特点,提出了一种基于RBF神经网络调参的滑模变结构控制器。滑模变结构控制器能够使高超声速飞行器稳定飞行,但在系统状态到达滑模面后会产生剧烈的抖振现象,不利于工程应用。RBF神经网络在一定条件下可以任意精度逼近非线性函数,且具有较强的自学习、自适应和自组织能力。将RBF神经网络与滑模变结构控制相结合,一定程度上能够消除滑模控制的抖振问题。在高超声速飞行器的巡航状态下,分别加入高度阶跃指令和速度阶跃指令进行了仿真。仿真结果表明,所设计的RBF神经网络滑模变结构控制器使高超声速飞行器在保证快速性、鲁棒性和抗干扰性的同时,克服了执行机构的抖振问题。
Aiming at the characteristics of highly non-linear hypersonic vehicle and strong coupling, a sliding mode variable structure controller based on RBF neural network is proposed. The sliding mode variable structure controller can make the hypersonic vehicle fly steadily, but it will produce the fierce buffeting phenomenon when the system state reaches the sliding mode surface, which is not good for engineering application. Under certain conditions, RBF neural network can approximate nonlinear function with arbitrary precision, and has strong self-learning, self-adapting and self-organizing ability. The combination of RBF neural network and sliding mode variable structure control can eliminate the chattering problem of sliding mode control to a certain extent. Under the cruise state of hypersonic vehicles, the simulation was carried out by adding altitude step command and speed step command separately. The simulation results show that the designed RBF neural network sliding mode variable structure controller can make the hypersonic vehicle overcome the buffeting problem of the actuator while ensuring fastness, robustness and anti-interference.