航空发动机滑动局部线性模型及控制  

Moving Locally Linear Model and Control of Aeroengine

作　　者：蒋安常 韩永健 张伟 周振华 周龙 杨蓓[2] JIANG An-chang;HAN Yong-jian;ZHANG Wei;ZHOU Zhen-hua;ZHOU Long;YANG Bei(AECC Guizhou Honglin Aero Engine Control Technology Co.,Ltd.,Guiyang 550009,China;School of Aircraft Engineering,Nanchang Hangkong University Nanchang,Nanchang 330063,China)

机构地区：[1]中国航发贵州红林航空动力控制科技有限公司,贵阳550000 [2]南昌航空大学飞行器工程学院,南昌330063

出　　处：《航空发动机》2022年第3期94-100,共7页Aeroengine

基　　金：国家自然科学基金(51766011)资助。

摘　　要：为了提高航空发动机的控制性能和控制系统的可靠性,提出一种基于粒子群算法的航空发动机局部小区域自适应滑动的线性模型建立方法和双闭环自适应PI控制方法。在辨识区间内,以发动机转子转速为状态变量,采用在线实测数据和粒子群算法的参数估计方法,使模型辨识参数按照设定区间大小自适应跟踪滑动,从而保证线性模型能够精确逼近发动机的非线性动态。通过分析发动机燃油调节器的工作特性,建立了燃油调节器计量活门和电液伺服阀的传递函数,并根据所构建的模型,设计了航空发动机转子转速和燃油流量双闭环自适应PI控制系统以实现对航空发动机的精确控制。结果表明:利用局部滑动自适应辨识计算机得到的数据与发动机稳态、动态试验数据相吻合,且双闭环仿真控制性能满足航空发动机工作性能要求,表明所提出的航空发动机局部线性建模方法和自适应PI控制器参数算法对提高发动机的控制性能是有效的。In order to improve the control performance and the reliability of control system of aeroengine,a linear model establish-ment method of aeroengine local small domain adaptive moving and a double closed-loop adaptive PI control method were proposed based on particle swarm optimization algorithm.In the identification domain,taking the engine rotational speed as the state variable,using the on-line measured data and the parameter estimation method of particle swarm optimization algorithm,the model identification parameters could adaptively track the moving according to the size of the set interval,so as to ensure that the linear model could accurately approach the non-linear dynamics of the engine.By analyzing the working characteristics of engine fuel regulator,the transfer functions of fuel regulator meter⁃ing valve and electro-hydraulic servo valve were established.According to the constructed model,the double closed-loop adaptive PI control system of aeroengine rotational speed and fuel flow was designed to realize the accurate control of aeroengine.The results show that the data calculated by the local moving adaptive identification model is consistent with the engine steady-state and dynamic test data,and the double closed-loop simulation control performance meets the requirements of aeroengine performance.It shows that the proposed aeroengine local linear modeling method and adaptive PI controller parameter algorithm are effective to improve the engine control performance.

关 键 词：滑动线性模型 粒子群优化 自适应PI控制 航空发动机 

分 类 号：V233.6[航空宇航科学与技术—航空宇航推进理论与工程]