一类基于FNN的非线性系统自适应控制

模糊神经网络用于控制,主要是为了解决复杂的非线性、不确定、不确知系统的控制问题。由于模糊神经网络具有学习能力和自适应性,使得其能对变化的环境有自适应性,控制器也基本上不依赖于模型,针对一类非线性系统,利用模糊神经网络对系统进行建模提出一种鲁棒自适应控制方法。首先利用李雅普诺夫定理证明在一定的条件下,闭环系统必能稳定,并证明这个条件即非线性函数f(x)中的x必落入某一紧集中成立,同时考虑其控制性能,选择鲁棒控制量,使跟踪误差达到要求的性能指标。理论分析和仿真结果说明了该控制算法的可行性和有效性。

FNN Adaptive Robust Control of Nonlinear Systems

Fuzzy neural network (FNN) applied in control mostly is in the nonlinear or uncertain systems for its self-study and adaption capacity. Its adaption to changing circumstance makes control rarely depend the model. For a class of unknown nonlinear systems, a FNN adaptive robust control scheme was presented. We first proved that the closed-loop system must be stable under a certain condition with Lyapunov theory, then proved the condition that the state x of the system falls into a compact set at all time. Moreover, considering the control performance, selected a robust control value to attenuate the effect of both the external disturbance and system approximation error to a prescribed level. Theory analysis and simulation results show the feasibility and effectiveness of the control algorithm.