【摘要】 为解决飞行数据中高频噪声对后续应用产生干扰的问题,提出基于自适应完备集合经验模态分解(CEEMDAN)和散布熵的飞行数据滤波方法。运用CEEMDAN将飞行数据自适应分解成一组固有模态函数(IMF);采用散布熵判断各IMF的噪声含量,若IMF中的有用信号完全被噪声掩盖,则予以去除,若IMF包含少量高频噪声,利用无偏似然估计阈值与改进阈值函数进行去噪;将处理后的IMF叠加以重构飞行数据。通过实例验证了该方法的有效性,其可提高飞行数据的可信度。更多还原

【Abstract】 To solve the problem that the high-frequency noise in flight data interfere with subsequent application,a flight data filtering method based on complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN)and dispersion entropy was proposed.The flight data were adaptively decomposed into a set of intrinsic mode functions(IMFs)by CEEMDAN.Dispersion entropy was used to identify the noise content of each IMF.If the useful signal in the IMF was completely covered by noise,it would be removed.If the IMF contained a little high-frequency noise,unbiased risk estimate threshold method and improved threshold function were chosen to remove high-frequency noise from IMF.The processed IMF was superimposed to reconstruct the flight data.The feasibility of the method which improved the reliability of flight data was validated by practical application.更多还原