多项式移动平滑算法及在试飞数据预处理中的应用  

A New Smoother Algorithm for Pre-Processing Flight Test Data

作　　者：杨亚军[1] 李言俊[1] 邓建华[1] 

机构地区：[1]西北工业大学

出　　处：《西北工业大学学报》1998年第4期501-505,共5页Journal of Northwestern Polytechnical University

基　　金：航空科学基金

摘　　要：提出了一种多项式移动平滑算法，并将其应用于带量测噪声的试飞数据的处理。该算法可进行状态估计、状态导数估计、量测噪声强度估计。整个算法公式建立在统一的矩阵形式基础上，因此算法中参数可以方便调整。并指明了在估计状态、状态导数、及噪声强度时，平滑多项式数据段的选取方法。在此基础上，还提出了一种新的野值处理法及流程。算例结果表明，该方法简单可行，能有效地完成数据预处理的若干任务。Flight test data, always contaminated by measurement noise, need to be pre - processed. For such pre -processing, we propose a new smoother aigorithm, called by us moving - polynomial smoother algorithm. Compared with existing pre - processing algorithms,our new algorithm is somewhat better in precision and more importantly, much simpler. For obtaining good precision, we explain how to select the data interval so as to guarantee that the estimate of noise intensity be unbiased. Our algorithm is simple because the mathematicai formulations are described by vectors and matrixes. Moving - polynomial algorithm can be used to estimate the true values of state variables and their derivatives and to estimate the intensity of the measurement noise. It can identify abnormal values and replace them with their estimated true values. We used moving -polynomial algorithm to pre - process some actual flight test data for angle of attack and the results confirm preliminarily the effectiveness of our method.

关 键 词：飞行试验 数据处理 多项式平滑算法 

分 类 号：V217.1[航空宇航科学与技术—航空宇航推进理论与工程]