弹丸协调臂的RBF神经网络自适应滑模控制

Adaptive Sliding Mode Control Based on RBF Neural Network for Projectile Transfer Arm

针对某大口径火炮弹丸协调臂电液伺服系统的位置控制问题,提出一种基于神经网络最小参数学习法的RBF网络自适应滑模控制方法。结合RBF神经网络具有局部逼近特性和神经网络最小参数学习法调节简单的优点,以电液伺服系统的状态为神经网络的输入,通过选取合适的参数,以神经网络的输出逼近系统的未知理想控制律。引入鲁棒项,保证控制策略的稳定性,并采用非线性函数调整反馈项参数的变化,保证收敛速度。仿真结果表明:控制算法在系统参数大范围变化的情况下能够保证弹丸协调臂的运动精度,并具有较好的鲁棒性。

Aiming at the position control problem of a large-caliber artillery projectile transfer arm electro-hydraulic servo system, an adaptive sliding mode control method based on RBF neural network minimum parameter learning method was proposed. Because the RBF neural network has local approximation characteristics, and the neural network minimum parameter learning method is simple to adjust, the state of the electro-hydraulic servo system was used as the input of the neural network. By selecting appropriate parameters, the output of the neural network approximates the unknown ideal control law of the system. Robust term was introduced to ensure the stability of the control strategy, and nonlinear function was used to adjust the changes of the feedback parameters to ensure the convergence speed. The simulation results show that the control algorithm can guarantee the motion accuracy of the projectile transfer arm and has better robustness under the condition that the system parameters vary widely.