摘要
为了预处理得到正确反映飞机飞行状态的飞行参数记录数据,介绍一种有效的用于飞行动作评估的飞参数据预处理方法。利用主成分分析相关性分析法提取与飞行动作最相关的几个参数数据,将各飞行参数的相关性融入多维自回归数据的预测估计过程,并对某型飞机的部分飞行参数数据进行实验分析,利用多维时间序列建模对其中5个参数的缺失数据进行一步预报得到估计值,与真实值进行比较得到的相对误差值都比较小(平均值0.03),该结果表明:该方法能得到正确反映和重现飞机飞行动作的时间序列数据,可推广到任意一种多元时间序列来表示目标状态的情况。
In order to get the record data of flight parameters correctly reflect the aircraft flight conditions by the means of preprocessing pretreatment, introduce an effective pretreatment method of flight parameter data for acrobatic maneuver evaluation. The several parameters which has higher relevance with acrobatic maneuver was extracted their data using the principal component analysis. And the correlation of the parameters was imported into the multidimensional auto-regression prediction data estimation process. We used a part of a certain aircraft flight parameters data for experimental analysis. In the experiment, the estimated data of 5 parameters＇ missing data by one step forecasting of multivariate time series modeling were compared with the true data, and the relative errors were all small that the average value was of 0.03. The results show that this method can get the time series data which can reflect or reproduce the acrobatic maneuver correctly. And the method can be extended to any kind of multivariate time series to represent the status of objectives.
出处
《兵工自动化》
2015年第5期22-25,共4页
Ordnance Industry Automation
关键词
飞行动作
飞行参数数据
主成分分析
多维时间序列建模
acrobatic maneuver
flight parameters data
principal component analysis
multivariate time series modeling
