论文部分内容阅读
控制对象参数的时变是噪声有源控制付诸实际应用所面临的主要问题之,传统的控制方法通常不考虑对象参数时变。本文首先引入一个能方便进行在线自适应的扩展控制对象自适应神经网络模型,在此基础上提出一种噪声有源控制的自适应神经网络方法。通过在控制过程中分别对控制网络和模型网络进行自适应,解决了控制对象参数的时变问题,显著改善了整个系统的鲁棒性。实验结果表明,对于控制对象参数的突变扰动,该方法具有良好的鲁棒稳定性。
The time-varying of the control object parameters is the main problem faced when the active noise control is put into practical application. The traditional control methods usually do not consider the time-varying of the object parameters. In this paper, an adaptive neural network model of extended control object that can be easily online adaptively is introduced. Based on this, an adaptive neural network method of active noise control is proposed. By adaptively controlling the control network and the model network in the control process, the time-varying problem of the control object parameters is solved, and the robustness of the entire system is significantly improved. The experimental results show that this method has good robust stability for the perturbed perturbations of the control object parameters.