基于一步最小二乘法建立航空发动机状态变量模型  

A Method of Establishing Aeroengine State Variable Model based on One-Step Least Square

作　　者：郑斐华 胡春艳 李伟[1,2] 郭磊 刘存喜[2] 

机构地区：[1]中国科学院大学,北京100190 [2]中国科学院工程热物理研究所轻型动力重点实验室,北京100190

出　　处：《燃气轮机技术》2017年第4期27-32,37,共7页Gas Turbine Technology

摘　　要：研究采用一步最小二乘法建立用于航空发动机控制系统及故障诊断设计中的状态变量模型,结合伪随机二进制激励信号求取模型矩阵系数。在线性最小二乘准则下,算法保证模型动态响应与非线性模型响应一致,过程一步完成无须迭代。利用一步最小二乘法建立了JT9D涡扇发动机稳态工作点的小偏差状态变量模型,仿真表明该方法得到的状态变量模型能够精确拟合发动机的时域和频域特性。相比与传统拟合方法,该方法不仅计算速度快,而且拟合精度大大提高,验证了方法的有效性。The one-step least square method was used to establish the state variable model （SVM） for the design of aeroengine controlsystem and fault diagnosis. Coefficients of SVM were obtained in the excitation signal： pseudo-random binary signal （PRBS）. Underthe linear least squares criterion,the algorithm ensures that the dynamic response of the model is consistent with that of the nonlinearmodel,and the whole process is without any iteration. The SVM of JTT 9D turbofan engine around the operating point of steady-state was established by one-step least square method. The simulation results show that the SVM obtained by this method can fit the time-domain and frequency-domain characteristics of the engine accurately. Compared with the traditional fitting method,not only the speed isfaster,but also the accuracy of the fitting model is greatly improved,and the validity of the method is verified.

关 键 词：状态变量模型 最小二乘法 伪随机二进制信号 航空发动机建模 

分 类 号：V233.7[航空宇航科学与技术—航空宇航推进理论与工程]