飞行/推进系统自适应神经网络综合控制仿真研究

 提出一种基于发动机喘振裕度自适应的飞行 /推进系统综合控制。在发动机喘振裕度较大的某些飞行条件或飞行包线内,通过调整喷口面积,使发动机喘振裕度保持在一个较小值,既保证发动机稳定工作,又增加发动机推力,从而改善飞机的性能。采用分散控制方案,综合控制系统由 5个控制子系统组成。各控制子系统的设计采用自适应控制和神经网络相结合的方法,所提出的参数和权重的自适应调整律保证系统的稳定性。全包线范围内飞机平飞加速和爬升数字仿真结果表明,该综合控制方法可缩短飞机的平飞加速时间和爬升时间。

SIMULATION STUDY OF ADAPTIVE NEURAL NETWORK CONTROL FOR INTEGRATED FLIGHT/PROPULSION SYSTEMS

An adaptive neural network control strategy is proposed for integrated flight/propulsion systems. In order to improve the performance of the flight/propulsion system, additional thrust is obtained by uptrimming the engine pressure ratio in some operation conditions and in part of the flight envelope where an excess surge margin exists. Using the decentralized control concepts, the integrated system consists of 5 controllers: rotor speed controller, high compressor surge margin controller, turbine exhaust temperature controller, pitch angle controller and altitude controller. Each subsystem is designed using adaptive control with neural network compensation. The stability of the system is guaranteed by the proposed parameter and weight turning law. Simulation results show that the flight acceleration and climb performances are improved using the proposed control.