奇异谱分解在超声速无人机声振试验数据处理中的应用

Singular value spectral decomposition and its application in acoustic vibration test data processing of a supersonic aircraft

将相空间重构和奇异谱分解相结合对受强气动噪声影响的超声速飞行器测试数据进行滤波,以实现对于试验参数的精确识别。首先通过数值仿真论证该方法的可行性,然后针对某型超声速无人机的声振试验对试验采集数据进行相空间重构,并对重构后的轨迹矩阵进行奇异值分解,得到反映真实信号信息的信号子空间和反映噪声信息的噪声子空间。通过定义奇异值差分谱这一指标来判定真实信号信息子空间维数,并针对现有最大差分谱理论缺陷提出了优选差分谱峰值理论,利用奇异谱分解的逆过程对真实信号进行重构。重构结果表明,该方法适用于超声速飞行下的飞行器声振试验数据处理,为超声速飞行器飞行状态的精确描述提供了良好的思路。

In order to realize precise identification of acoustic vibration test data of a supersonic aircraft,a new filtering method combining phase space reconstruction and singular spectral decomposition was proposed.Firstly,the feasibility of this method was demonstrated through numerical simulation.Secondly,in order to separate the signal subspace and the noise subspace,the phase space reconstruction of the test data was conducted,and the attractor track matrix was also decomposed with singular value decomposition （SVD ）.Finally,aiming at shortages of the maximum difference spectrum theory, the concept of optimizing difference spectrum theory was presented, and the signal reconstruction was proposed on the basis of the peak position of the optimizing difference spectrum.Reconstruction results showed that the proposed method is suitable for processing the acoustic vibration test data of a supersonic aircraft,the result provided a good foundation for the precise description of a supersonic aircraft＇s flying state.