论文部分内容阅读
为提高多变量系统解耦控制的性能,提出了一种基于参考模型的神经网络在线解耦控制方法.构造神经网络实现前馈解耦,通过参考模型的输出与被控系统输出估计耦合作用对被控量的影响,由此设计神经网络权值参数学习算法,在线调整网络参数使多变量耦合系统实现解耦;对解耦后的子系统分别设计闭环控制器,以达到优良的控制性能.仿真实验结果表明,提出的解耦控制方法是简单有效的.
In order to improve the performance of multivariable decoupling control system, a neural network online decoupling control method based on reference model is proposed.The feedforward decoupling is constructed by neural network, and the coupling effect between the output of the reference model and the output of the controlled system Which is influenced by the amount of control. Based on this, the learning algorithm of the neural network weight parameter is designed and the network parameters are adjusted online to decouple the multivariable coupling system. The closed-loop controllers are respectively designed for the decoupled subsystems in order to achieve excellent control performance. Simulation results show that the proposed decoupling control method is simple and effective.