非线性MIMO系统神经网络自适应方法及其在飞控中的应用.pdf

18 2 Vol. 18 No. 2 1997 3 A TA AERONAU TI A ET ASTRONAUTI A SINI A M ar. 1997 MIMO ( , ,710072) NEURAL-NETWORK BASED ADAPTIVE METHOD FOR MIMO NONLINEAR SYSTEM AND ITS APPLICATION TO FLIGHT CONTROL Zhou Xuan, Shi Zhongke ( Department of Automatic ontrol,Northwestern Polytechnical University, Xian, 710072) M IMO , , BP , BP 273. 1, 249. 1 TP V Abstract A method is presented for controlling unknow n nonlinear MIMO systems by using diagonal recurrent neural network DRNN( ) . In this controlling scheme unknow n nonlinear, plant is identified by a diagonal recurrent neuroidentifier ( DRNI) , and the sensitivity informa- tion of the plant is provided to a diagonal recurrent neurocontroller(NRN ) . A dynamic back- propagation algorithm DBP is used to train DRN and DRNI The proposed approach is ap-( ) . plied to a flight control system and the simulation results are included. Key w rdsneural netw ork, adaptive control system , dynamic backpropagation algorithm , (MNN) (RBFN) [ 1] - (SISO) , [ 2] - (MIMO) , MIMO MIMO SISO , , 1 y i (k + di) = Fi [y (k) , ,y (k - p y + 1) ,u(k) , ,u(k - p u + 1) ] ( 1) T T 1 2 M 1 2 N :y (k) = [y (k ) ,y (k ) , ,y (k ) ] , u(k ) = [u (k) , u ( k) , , u (k) ] 1996-02-13 , 1996-07-30 164