摘要
对于现代飞行控制系统,提出用一种基于Sugeno型模糊推理的自适应神经网络模糊系统ANFIS代替传统的根据马赫数或高度单参数调参的控制律,通过设计和训练得到的模糊系统可实现根据高度和马赫数的双参数增益调参.仿真结果验证了该模糊系统可以保证训练点上输出与原设计相同的最优反馈增益,同时泛化出平衡点之间的增益参数,且当系统纵横向存在较大建模误差时同样也有较好的控制效果.
An ANFIS based on Sugeno-type fuzzy inference is developed for modern flight control system. It can replace the traditional gain scheduling according to the altitude or Mach number which is only one-parameter adjusting. The ANFIS trained by the designed data can be used to realize double parameters gain regulating. The simulation results verify that the ANFIS can generate the same gains at the designed flight conditions. It can get a good control effect on other untrained flight conditions, also when the system has huge modeling error.
出处
《复旦学报（自然科学版）》
CAS
CSCD
北大核心
2010年第4期530-534,共5页
Journal of Fudan University：Natural Science
作者简介
颜振萍（1984-），男，硕士研究生，E-mail：yanzhenping@gmail．com．
