基于径向基函数网络的非线性通用模型控制

为了克服通用模型控制器要求过程一阶微分模型应该有显式解的局限性，提出了一种基于神经网络的通用模型控制方法，将非线性过程模型应用逆系统的方法在控制算法中直接嵌入过程模型，从而保证通用模型控制策略的可实现性。其参考轨迹是一条典型的二阶曲线，由于径向基函数网络具有许多优点，该控制策略中的神经网络为径向基函数网络。该控制器参数具有明显的物理意义，参数整定方便。仿真实验验证了该控制策略的有效性。

Nonlinear Common Model Control Based on Radial Basis Function Neural Networks

In order to overcome the weakness of explicitly calculating the manipulated input according to the first-order differential equation about the controlled plant in the common model control(CMC) scheme,we propose a common model control method based on neural networks,which can embed the process model into the controller by the inverted control method with neural network.It can guarantee the(realizability) of the common model control scheme based on neural networks.The reference trajectory is a classic second-order curve.As the radial basis function is much superior in control system,the neural network in the control scheme is the radial basis function neural network(RBFNN).The parameters of CMC based on neural network have very explicit physics meaning and it is very easy to tune for the controller.The simulation results show the effectiveness of the nonlinear common model controller based on RBFNN.