基于DRNN网络的航空发动机多变量解耦控制

通过分析对角递归神经网络（DRNN）及带动量项的梯度学习方法（GDM）,针对某型涡扇发动机的性能控制,研究了基于对角递归神经网络的多变量自学习解耦控制算法及其在航空发动机控制中的应用.阐明了该方法的结构和原理.并在设计点处进行了发动机多变量解耦控制系统设计.在偏离设计点时,大量的仿真结果表明,系统具有较好解耦和自适应能力.

A multivariable decoupling control based on DRNN for aeroengine

A new neural paradigm called diagonal recurrent neural network(DRNN) and Gradient Decent Method(GDM) were presented.A multivariable decoupling control algorithm based on the diagonal recurrent neural network was used for aeroengine control.A generalized GDM was developed and used to train diagonal recurrent neuroidentifier.The emphasis was focused on the research of the algorithm and the properties of the controller,as well as their application to aeroengine control by computer simulation.Finally,the aeroengine control system based on DRNN was designed.Simulation shows that the system has fine performance of decoupling and adaptive capabilities.