摘要
讨论了模糊逻辑控制与神经网络相结合的一种控制方法,给出了一种增益自适应调整的模糊控制方法和BP网络自适应变步长学习算法,前者提高了系统对参数变化的适应能力,同时也提高了系统的控制精度,改善了系统品质,后者缩短了网络的学习时间,有利于实时控制的实现。这两种方法成功地用于直升机飞控系统的设计。同时。
A discussion is devoted to the intelligent control combining fuzzy logical control with artificial neural networks. A new kind of the intelligent control system is presented, which regulates the gain of fuzzy control and the step length of learning algorithm of BP networks adaptively. The regulation of fuzzy control gain increases the adaptive ability for the parameters of system as well as the precision, and improves the qualities of control system. The regulation of the step length of the learning algorithm of BP networks shortens learning time of the network. These benefit the realization of real time control. The method has been applied to design of a helicopter flight control system. The digital simulation for the helicopter has been done and the results demonstrate that the present method is advanced and useful.
出处
《南京航空航天大学学报》
EI
CAS
CSCD
北大核心
1999年第2期185-191,共7页
Journal of Nanjing University of Aeronautics & Astronautics
基金
国防科技重点实验室基金