磁悬浮球系统的神经网络辨识及变结构控制

磁悬浮球系统是一个典型的非线性的不稳定的系统,基于对其建模的复杂性和不准确性,文章利用神经网络能逼近任意非线性函数这一特性,对磁悬浮球系统进行辨识;再根据滑模变结构控制原理设计了磁悬浮球系统的变结构控制器,利用MATLAB对系统进行建模仿真,仿真结果表明,RBF网络能很好地逼近本磁悬浮球系统;滑模变结构控制对于此非线性系统有较好的控制效果,小球能很快地悬浮在平衡位置,该控制系统具有较好的稳态特性和抗干扰性.

Identification Based on Neural Network and Variable Structure Control of a Magnetic Levitation System

Magnetic levitation system is a typical nonlinear and instable system. Based on the complexity and inaccuracy of modelling, this paper identifies magnetic levitation system through the speciality that neural network can approach any nonlinear function. A variable structure controller is designed in line with the variable structure control principle. Besides, a simulation model of the system is proposed by using MATLAB, and the result shows that RBF network could approach magnetic levitation system very well; variable structure controller has a good effect on this nonlinear system and the ball could suspending at balance position very quickly; this control system has a preferable- stability and anti-disturbance property.