论文部分内容阅读
近空间可变翼飞行器具有可伸缩小翼结构,针对小翼伸出和收回状态变化过程中,存在参数摄动,可能造成飞行器状态以及控制量的大范围跳变,影响飞行器的稳定性的问题,本文提出一种快速双幂次趋近律滑模与神经网络结合的自适应滑模控制方法。应用该方法设计快速双幂次趋近律滑模控制器,并利用神经网络充分逼近复杂的非线性关系能力,得到小翼伸缩全过程的滑模趋近律。对比分析传统滑模控制方法和快速双幂次滑模与神经网络结合自适应控制效果,仿真结果表明快速双幂次滑模控制与神经网络结合方法具有良好的控制效果。
The near-space variable-wing aircraft has a retractable winglet structure. In the process of projecting and retracting the winglets, there are parameters perturbations that may cause large-scale changes of the aircraft state and control amount and affect the stability of the aircraft , This paper presents an adaptive sliding mode control method combined with fast bimodulus approach law and neural network. Applying this method, a fast barycenter approach law sliding mode controller is designed. By using the neural network, the complex non-linear relationship ability is sufficiently approximated to obtain the sliding-mode approach law of the whole winglets. The traditional sliding mode control method and fast bivariate sliding mode and neural network are compared and analyzed. The simulation results show that the combination of fast bivariate sliding mode control and neural network has a good control effect.