摘要
通过分析飞行记录参数奇异点的特点及产生原因,建立了奇异点在某一时间区域内连续信号的数学描述。在软件设计中应用奈奎斯特定律确定数据段窗口的大小、Matlab仿真技术、多项式和最小二乘法,检测偏移量是否超出规定阈值,提出了飞参系统记录数据的奇异值预处理的方法及软件设计思想;通过多机种多架次的大量飞行记录数据的实验,验证了采用多项式和最小二乘法是对飞参系统记录数据奇异值进行预处理的一种有效方法,对其他领域信号处理具有普遍适用性。
The paper establishes the continuous signal mathematical description of flight data strange points in one time domain by analyzing its features and causes. We Offset detection threshold exceeding the prescribed limit and put forward the method and software design process by polynomial fitting and the application of Nyquist＇s law to determine the size of the data segment window, Least-squares method and MATLAB simulation technique. Through large amounts of aeroengine recorded flight data experiment, is shows the polynomial fitting and least-squares method is effective for the pretreatment of strange points. Also it acquires a very good result and has widespread adaptability to other parameters dealing.
出处
《电机与控制学报》
EI
CSCD
北大核心
2009年第A01期135-138,共4页
Electric Machines and Control
关键词
奇异点
多项式
最小二乘法
仿真
strange point
polynomial fitting
least-squares method
simulation
