摘要
虽然滑模控制具有控制简单和对不确定性与扰动不灵敏等优点,但是控制信号中的颤动是其应用中需解决的主要问题。该文首先针对一类非线性系统提出了一个新型控制器—模糊神经网络滑模控制器。新控制器不仅能消除颤动,而且比一般滑模控制器具有更强的鲁棒性。然而它与一般滑模控制器相比有较大的跟踪误差。为了解决这个问题,提出了结合滑模控制器和模糊神经网络滑模控制器的自适应控制方法。这种自适应控制方案可以减小跟踪误差,增强系统的鲁棒性和消除控制信号中的颤动。仿真结果说明了控制方案的有效性。
Sliding mode control is simple and insensitive to uncertainties and disturbances, but control input chattering is the main problem in the sliding mode controller (SMC). A new type controller-fuzzy neural networks sliding mode controller (FNNSMC) is presented firstly for a class of nonlinear systems. The FNNSMC can eliminate the chattering indeed and is robuster than the SMC while the bounds of the uncertainties and the disturbances of the system are unknown. However there has larger tracking error in the FNNSMC than that in the SMC. To overcome the problem, an adaptive controller, where the FNNSMC and the FNN are incorporated by the smooth transformation, is studied. This adaptive control scheme can decrease the tracking error, enhance the system抯 robustness and eliminate the high frequency chattering in the control signal. The simulation results demonstrate the advantages of the algorithm.
出处
《中国电机工程学报》
EI
CSCD
北大核心
2002年第5期78-83,共6页
Proceedings of the CSEE
关键词
模糊神经网络
滑模控制器
非线性系统
自适应控制
nonlinear systems
fuzzy neural networks
sliding mode control
adaptive control
