非线性动态系统的模糊神经网络自适应H_∞鲁棒控制

摘要

To deal with the inherit instability of non-linear dynamic systems, a direct adaptive control method for an robust fuzzy neural network based nonlinear dynamic system controller was developed. The controller is composed of an control element for the nonlinear dynamic system and a direct adaptive control element. On-line learning of the fuzzy neural network is used to regulate errors caused by modeling uncertainties in order to guarantee system stability and performance. The stability of the designed system was proven in this paper, and a simulation demonstrated the effectiveness of the proposed method.%针对非线性动态系统不稳定性,提出了一种基于模糊神经网络的H_∞鲁棒直接自适应控制方法,通过设计非线性动态系统的H_∞控制器,采用模糊神经网络在线学习对动态系统的建模不确定性产生的误差,可保证不确定闭环稳定并具有H_∞性能,并证明了模糊神经网络H_∞鲁棒直接自适应控制系统的稳定性,仿真算例也表明了该方法具有较好的抗干扰性能、鲁棒性较好.