基于自适应无迹卡尔曼滤波QAR数据降噪研究

Study on Noise Reduction of QAR Data Based on Adaptive Unscented Kalman Filter

针对QAR数据包含离群值、噪声值等异常数据严重影响数据分析的问题,提出了一种自适应无迹卡尔曼滤波的数据降噪方法。利用改进拉依达准则剔除粗大误差数据,以无迹卡尔曼滤波为基础,结合Sage-Husa噪声估计器对系统噪声进行实时预测和修正,有效地解决了系统噪声时变的问题。利用空客A330飞机的数据样本对算法有效性进行了数值验证,仿真结果表明,自适应无迹卡尔曼滤波算法估计精度更高,降噪效果更优。研究可提高基于QAR数据分析与挖掘工作的数据质量。

Aim to the problem that QAR data contain outliers,noise values and other abnormal data,which seriously affect the data analysis,an Adaptive Unscented Kalman Filter(AUKF)method for QAR data noise reduction is proposed.Using the improved Pauta criterion to eliminate the gross error data,the system noise was predicted and corrected in real time based on the Unscented Kalman Filter(UKF)and the Sage-Husa noise estimator,which effectively solved the time-varying problem of system noise.The validity of the algorithm was verified by using the data sample of airbus A330 aircraft.The simulation results show that the adaptive unscented Kalman filter algorithm has higher estimation accuracy and better noise reduction effect.The research can improve the data quality based on QAR data analysis and mining.