利用飞行试验信号对发动机模型辨识的研究

 利用飞行试验信号对发动机模型辨识进行了研究。在飞行过程中,发动机将受到各种随机扰动的作用,这种扰动将引起发动机特性的变化、在这种情况下,利用与发动机特性有关的信号进行发动机数学模型辨识对发动机自适应控制和状态监控有实际的意义。飞行试验是在H=3km、v=770km/h和H=13km、v=1540km/h时进行的,利用机载信号采集与记录设备录取了试验信号。介绍了信号的平滑和滤波方法,讨论了巴特沃思滤波器的设计和参数选择。为了得到无偏估计,采用递推广义最小二乘方法进行辨识,并得到了辨识结果。

IDENTIFICATION INVESTIGATION OF ENGINE MODEL USING FLIGHT TEST SIGNALS

The engine model identification using flight test signals is introduced. As the aircraft flights, the engine will be subjected to random disturbances, which will result in change of the engine performance. In this case, the engine model identification using flight text signals concerned with the engine performance will have practical meaning for adaptive control and condition monitoring of the engine.The flight tests are accomplished at the altitude H = 3km, flight speed v = 770km / h and at H= 13km, v= 1540km/h. The test signals are recorded by on-board signal sampling and recording equipment. The signal smoothing and filtering methods are introduced. Design and parameter selection of Butterworth filter are discussed.In order to obtain invariant estimation of the model parameters, the recursive generalized least square identification method is used the results are given.